| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology (P.A., V.V., G.B., B. Taouli, Y.M.), Pathology (V.P., B. Terris), and Digestive Surgery and Transplantation Unit (J.B.), Hopital Beaujon, Clichy, France. From the 1999 RSNA scientific assembly. Received February 12, 2002; revision requested April 10; final revision received November 17; accepted December 16. Address correspondence to G.B., Department of Radiology, University of Palermo, via Villaermosa 29, 90139 Palermo, Italy (e-mail: gbranca@yahoo.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: US, helical multiphasic CT, and MR images in 13 patients with pathologically proven telangiectatic FNH were reviewed retrospectively. Two abdominal radiologists evaluated lesions for number, size, heterogeneity, surface characteristics, presence of a central scar, presence of a pseudocapsule, US appearance, attenuation at CT, signal intensity at MR imaging, and presence of associated lesions. Imaging and pathologic findings were compared.
RESULTS: Sixty-one lesions (5–140 mm in diameter) were seen at imaging. Lesions were multiple in eight of 13 (62%) patients. Imaging characteristics were heterogeneity in 26 of 61 lesions (43%), well-defined margins in 43 of 61 (70%), lack of a central scar in 56 of 61 (92%), presence of a pseudocapsule in three of 61 (5%), hyperintensity on T1-weighted MR images in 17 of 32 (53%), strong hyperintensity on T2-weighted MR images in 24 of 54 (44%), and persistent enhancement on delayed contrast material–enhanced CT or T1-weighted MR images in 23 of 38 (61%). No specific US pattern was noted. Two patients had additional lesions: One had classic FNH, and the other had a cavernous hemangioma. Hyperintensity on T1-weighted MR images was due to sinusoidal dilatation. Hyperintensity on T2-weighted MR images correlated well with the presence of inflammation.
CONCLUSION: Telangiectatic FNH differs from typical FNH at imaging: Atypical FNH features often observed with telangiectatic FNH are lack of a central scar, lesion heterogeneity, hyperintensity on T1-weighted MR images, strong hyperintensity on T2-weighted MR images, and persistent contrast enhancement on delayed contrast-enhanced CT or T1-weighted MR images.
© RSNA, 2003
Index terms: Liver, focal nodular hyperplasia, 761.3198 • Liver neoplasms, CT, 761.12111, 761.12112, 761.12114 • Liver neoplasms, MR, 761.121411, 761.121412, 761.121415, 761.12143 • Liver neoplasms, US, 761.1298
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Unfortunately, when one or more of these signs are lacking at imaging, a confident diagnosis is not always rendered. Furthermore, sometimes atypical findings can be observed in FNH, such as multiplicity, lesion heterogeneity, fatty infiltration, nonvisualization of a central scar, presence of a pseudocapsule, diameter larger than 10 cm, absence of contrast enhancement of the scar at delayed imaging, and hypointensity of the scar at T2-weighted MR imaging (6,7).
Besides the imaging appearances of previously reported atypical cases, different histologic forms have also been described at pathologic examination (8,9). Wanless et al (8) and Nguyen et al (9) reported the pathologic features of telangiectatic FNH, characterized by one-cell-thick hepatic plates separated by sinusoidal dilatation, with no central scar and no architectural distortion (9). In these two studies, the frequency of patients with telangiectatic FNH ranged from 9.5% to 19% (8,9). According to these authors, there are two differences between classic and telangiectatic FNH: (a) In telangiectatic FNH, arteries have hypertrophied muscular media but no intimal proliferation in contrast to the classic form. (b) In telangiectatic FNH, these abnormal vessels drain directly into the adjacent sinusoids, while in classic FNH, connections to the sinusoids are almost never seen (8,9).
To our knowledge, no series of telangiectatic FNH has been described in the radiology literature. We hypothesized that imaging features of telangiectatic FNH might differ from those commonly observed in classic FNH. Thus, the purpose of our study was to review the US, helical multiphasic CT, and MR imaging findings in 13 patients with telangiectatic FNH and to compare imaging features with histopathologic results from resected specimens.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
All patients were women aged 26–50 years (mean age, 38 years), and all had a history of oral contraceptive intake for a mean duration of 15 years (range, 6–30 years). Nine patients were asymptomatic (the lesions were discovered fortuitously), three patients had abdominal pain (chronic pain [n = 2] or acute pain [n = 1] due to intratumoral bleeding with no subcapsular or intraperitoneal hematoma), and one patient had chronic fever and asthenia. Five patients had normal liver test results. Three had increased serum 
-glutamyltransferase levels (two to 12 times the normal upper limit). One had isolated increased serum alkaline phosphatase activity (three times the normal upper limit). Four had increased serum -glutamyltransferase levels (between two and eight times the normal upper limit) and increased serum alkaline phosphatase activity (between 1
and 3 times the normal upper limit).
Surgical resection was performed in all patients and consisted of six major liver resections (involving more than three segments) and seven segmentectomies or wedge resections. In patients with multiple lesions, resection of one or more lesions was associated with biopsy of at least one remaining nodule. None of the patients had underlying hepatic cirrhosis, and none had elevated serum 
-fetoprotein levels.
Imaging Protocols
All patients underwent ultrasonography (US), helical multiphasic CT, and MR imaging at our institution. This is our usual strategy in patients with liver tumors and equivocal features. US was the first modality used, followed by either CT or MR imaging. All examinations were performed within 30 days preceding surgery.
US was performed with two scanners (128XP/10, Acuson, Mountain View, Calif; or Elegra, Siemens, Erlangen, Germany), with 2–5-MHz multifrequency curvilinear transducers and transducer frequencies selected to optimize imaging of the liver. Color or power Doppler US was not performed.
Helical multiphasic CT scanning was performed in all patients by using a CT Twin Flash unit (Marconi Medical Systems, Cleveland, Ohio) with a double-detector array and a scanning time of 1 second for 360° rotation, with 5-mm contiguous sections and a pitch of 1. After undergoing unenhanced acquisitions of the liver, all patients underwent helical multiphase CT that included both hepatic arterial phase and portal venous phase imaging 25–30 seconds and 60–70 seconds, respectively, after intravenous infusion of 140 mL of nonionic contrast material (350 mg of iodine per milliliter iohexol [Omnipaque]; Nycomed-Amersham, Cork, Ireland). Contrast material was administered at a rate of 4 mL/sec with a mechanical power injector (Medrad, Pittsburgh, Pa). Delayed-phase imaging (5–10 minutes after contrast medium injection) was performed in five patients.
MR imaging was performed in 10 patients with a 2.0-T Gyrex (Elscint, Haifa, Israel) superconducting system with a body coil. Sequences included fast spin-echo T2-weighted MR imaging (repetition time msec/echo time msec, 3,250/110; matrix, 200 x 256; echo train length, 20; field of view, 400 mm; section thickness, 10 mm; gap, 
2 mm; two signals acquired) and T1-weighted gradient-echo MR imaging (160/4.9 [in phase]; matrix, 200 x 256; flip angle, 100°; field of view, 34 cm; section thickness, 10 mm; gap, 2 mm; two signals acquired).
Three patients were imaged with a 1.5-T system (Gyroscan Intera; Philips Medical Systems, Best, the Netherlands) with a maximum gradient strength of 40 mT/m and a slew rate of 200 msec by using multiarray torso coils for signal reception. Pulse sequences included breath-hold T1-weighted fast field-echo sequences (216/5.1; flip angle, 80°; field of view, 34 cm; matrix, 200 x 256; number of sections, 24; section thickness, 8 mm; one signal acquired) and respiratory-triggered T2-weighted fat-suppressed turbo spin-echo sequences (1,600/70; flip angle, 90°; field of view, 34 cm; reconstruction matrix, 512 x 512; number of sections, 24; section thickness, 8 mm; two signals acquired).
In all patients, a gradient-echo T1-weighted MR sequence was performed during the hepatic arterial and portal venous phases (at 20 and 50 seconds, respectively) after manual administration of 0.1 mmol per kilogram gadoterate meglumine (Dotarem; Laboratoire Guerbet, Roissy, France), followed by a 20-mL saline flush. Delayed-phase imaging (5–10 minutes after contrast medium injection) was performed in seven patients.
Image Analysis
All images were interpreted retrospectively and jointly in consensus by two abdominal radiologists (P.A., V.V.; 5 and 15 years of experience, respectively) who had knowledge of the diagnosis of telangiectatic FNH but did not have knowledge of the specific number of tumors or clinicopathologic findings in any patient. CT and MR images obtained in each patient were reviewed separately. The total number of lesions in each patient was evaluated according to image findings from all modalities considered together.
US, helical multiphasic CT, and MR images were reviewed in each patient to analyze the following imaging criteria: (a) number of lesions; (b) lesion location according to the hepatic segment numbering system of Couinaud; (c) lesion diameter; (d) US pattern, classified as hypoechoic, isoechoic, or hyperechoic to the adjacent liver parenchyma; (e) attenuation at unenhanced and contrast-enhanced CT, classified as hypoattenuating, isoattenuating, or hyperattenuating to the adjacent liver parenchyma; (f) signal intensity characteristics of the lesions at unenhanced and contrast-enhanced T1-weighted MR imaging with regard to the surrounding liver parenchyma (hyperintensity on T2-weighted MR images was defined as slight if it was between liver and spleen intensity or strong if it was equal to or greater than that of the spleen); (g) homogeneous or heterogeneous appearance; (h) lesion surface, classified as smooth (round and regular) or lobulated (irregular and cloudy); (i) presence of a central scar; (j) presence of a pseudocapsule; (k) presence of a central artery. A central scar was defined as an area of distinctly different attenuation or signal intensity in or near the center of the lesion on nonenhanced images or on images obtained at different phases of enhancement. A pseudocapsule was defined as an enhanced rim on delayed-phase CT or MR images due to the presence of draining vessels or as an incomplete rim of variable thickness on CT or MR images. These findings were used to compare the imaging and gross pathologic features of the lesions.
We also recorded any imaging and histopathologic evidence of hepatic vascular abnormalities or associated lesions.
Statistical Analysis
Because multiple lesions were present in each patient, differences (size of lesions with and without necrosis, size of lesions with and without hemorrhage) were evaluated with a general linear model to eliminate dependency. The unit of measure in our statistical analysis was number of lesions, rather than number of patients.
Pathologic Analysis
All cases were reviewed retrospectively by one pathologist (B.T.) with expertise in hepatic pathology. The gross features of each resected tumor were analyzed with use of photomacrographs and included the number of lesions, their location, size, and contour, as well as congestion, hemorrhage, necrosis, and the presence of a macroscopic scar. Microscopic examination was performed on paraffin-embedded representative sections of the lesions, which were routinely processed and stained with hematoxylin-eosin stain, and cytologic aspects of hepatocytes were noted. The presence of a central scar was assessed. The following features were semiquantitatively evaluated in the lesions: steatosis, necrosis, hemorrhage, and sinusoidal dilatation.
To perform a correlation between sinusoidal dilatation and signal intensity, we used a three-category scale for the amount of sinusoidal dilatation—marked, intermediate, or mild, depending on the number and size of dilated sinusoids seen at histologic examination. A direct comparison between imaging and histologic findings was performed in resected lesions together by the pathologist and one radiologist (V.V.).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Imaging Findings
A total of 61 lesions were seen at imaging, and the number of tumors observed at US, CT, and MR imaging was 42, 46, and 58, respectively. All liver segments were involved. In five patients, lesions were solitary at imaging, with a mean diameter of 58 mm (range, 25–140 mm). Solitary lesions at imaging were also solitary at pathologic examination. In eight patients, lesions were multiple (two to 12 lesions per patient), with a mean diameter of 28 mm (range, 5–120 mm) (Fig 1). Table 1 shows the imaging characteristics of the lesions. Among the 54 lesions detected on T2-weighted MR images, 26 were slightly hyperintense, and 24 were strongly hyperintense (Fig 2). Lesion heterogeneity was detected with at least one imaging modality in 26 of 61 lesions (43%) (Fig 3). Two lesions in one patient with acute bleeding showed heterogeneous hyperintensity on both T1- and T2-weighted MR images. A pseudocapsule was detected in two patients for a total of three lesions (5%). In two lesions, the pseudocapsule was seen as an enhanced rim on delayed-phase MR images, while in one lesion, it was seen as an incomplete rim of variable thickness that surrounded the lesion.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Radiologic-pathologic correlation of the three lesions surrounded by a pseudocapsule at imaging showed compressed liver parenchyma (n = 1) and enlarged draining veins (n = 2). The five lesions that were shown to have a central scar at CT and/or MR imaging contained only large puddles of centrally located blood that separated hepatocyte nodules. In five patients, not all lesions were resected. All patients underwent intraoperative biopsy of at least one remaining lesion. In all cases, findings at pathologic examination confirmed the diagnosis of telangiectatic FNH. Thus, a total of 76 lesions were confirmed at pathologic examination (66 were resected, and 10 were sampled for intraoperative biopsy). Twenty-one lesions seen at imaging were not histologically proven, representing 32% of the total number of lesions seen at imaging.
Follow-up imaging (mean follow-up time, 27 months; range, 16–48 months) was performed in the five patients in whom not all lesions were resected. None of them had an increase in the size or number of nodules. The adjacent liver parenchyma was also studied in the resected specimens, indicating that five patients had steatosis that varyied from 10% to 60%. Percutaneous biopsy performed prior to surgery in seven patients (core-needle biopsy with an 18-gauge needle) yielded inconclusive results in two cases because of the paucicellularity of the specimen. In two patients with telangiectatic FNH, incorrect diagnoses of adenoma and FNH, respectively, were initially rendered at percutaneous biopsy. In the last patient, the pathologist could not discriminate hepatic adenoma from FNH.
Associated Lesions
Of the 13 patients with telangiectatic FNH, two (15%) had at least one other benign hepatic tumor. One patient with multiple telangiectatic FNH had one pathologically proven typical FNH tumor that was 4 cm in diameter and showed isoattenuation and isointensity on nonhenanced portal venous phase and delayed-phase CT and MR images, respectively, as well as strong homogeneous enhancement on arterial dominant phase images and slight hyperintensity on T2-weighted MR images. One patient with one telangiectatic FNH tumor had an associated cavernous hemangioma. The hemangioma was isoattenuating to the blood pool on nonenhanced and contrast-enhanced CT scans and showed nodular peripheral enhancement.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The female-male ratio of patients with FNH has been reported to be approximately 8:1 (9), while 100% of our patients were women. All patients in our current series and other series have been adults with a mean age of 38 years (9).
All of our patients with telangiectatic FNH had taken oral contraceptives for a longer period (mean time, 15 years) than that reported in patients with typical FNH (10).
Wanless et al (8) described telangiectatic FNH in five patients from a population of 27 (19%) with FNH who died and underwent autopsy. Three of these five patients had multiple lesions with a total of two, 12, and 16 lesions, respectively, while in the remaining two patients, the tumors were associated with typical FNH. The size of the lesions ranged from 2 to 3 cm. This association is probably not fortuitous, as we also observed the association of multiple telangiectatic FNH with typical FNH (n = 1) and with cavernous hemangioma (n = 1). More recently, an extensive pathologic study (9) showed that the frequency of patients with telangiectatic FNH was 9.5% (16 of 168 patients), and the frequency of telangiectatic FNH lesions was 15.4% (47 of 305 lesions). In these two series, the frequency of telangiectatic FNH seems higher in patients with multiple lesions than in those with solitary lesions. In our surgical population, patients with the telangiectatic variant represented 11% of the patients with FNH, a similar percentage to those in the two previous studies. Moreover, we found that telangiectatic FNH was multiple in almost half of the patients—a higher percentage than the ones recently reported in two large series of patients with FNH (23% and 21.6%, respectively [4,11]).
To our knowledge, only two case reports of telangiectatic FNH have been described in the English radiology literature. Peterfy and Rosenthall (12) described a 9-cm lesion with heterogeneous enhancement at contrast-enhanced CT. Haber et al (13) reported telangiectatic FNH with multiple lesions occurring in a 22-year old woman with Klippel-Trenaunay-Weber syndrome, with none of the lesions showing a central scar (13).
While US depicted 69% of the total lesions seen at imaging in the present study, it proved to be nonspecific for characterization of telangiectatic FNH. A common finding of telangiectatic FNH in our study was strong arterial enhancement, which is also seen in most cases of typical FNH (5). However, we observed other imaging characteristics in our patients: (a) a heterogeneous pattern in 26 of 61 (43%) lesions, (b) hyperintensity on T1-weighted MR images in 17 of 32 (53%) lesions, (c) strong hyperintensity on T2-weighted MR images in 24 of 54 (44%) lesions, (d) absence of a central scar in 56 of 61 (92%) lesions, and (e) persistent enhancement on delayed-phase contrast-enhanced CT or T1-weighted MR images in 23 of 32 (61%) lesions.
A heterogeneous pattern is a very rare feature in FNH and was observed in 2.4% of the cases in a radiologic-pathologic study by Vilgrain et al (11). In our series, all heterogeneous lesions at imaging were confirmed to be heterogeneous at pathologic examination. The main causes were necrosis, the degree of sinusoidal dilatation, and the presence of hemorrhagic foci. Our results show that hemorrhage, an unusual finding in FNH, was more commonly observed in larger lesions. Three patients with homogeneous lesions had no hemorrhage or necrosis at pathologic examination.
Hyperintensity on T1-weighted MR images is very rare in FNH and was observed in 2.1%–6.0% of cases in previous studies (11,14). In our series, lesion hyperintensity was observed in 53% of lesions on T1-weighted MR images, and at least one hyperintense lesion on T1-weighted MR images was detected in 62% of patients. It is well known that hyperintensity on T1-weighted MR images may be due to different pathologic changes, including fat deposition, copper accumulation, high protein concentrations, blood degradation products, or sinusoidal dilatation (15). In our patients, histopathologic analysis was performed in 12 of the 17 hyperintense lesions. There was no fatty infiltration and no hemorrhage in four lesions. Minimal fatty infiltration or minimal foci of hemorrhage was seen in eight of the 12 lesions. Copper accumulation or other causes of hyperintensity on T1-weighted MR images were not found. These findings suggest that lesion hyperintensity on T1-weighted MR images may reflect intrasinusoidal dilatation. However, semiquantitative evaluation results of intrasinusoidal dilatation were not well correlated with hyperintensity on T1-weighted MR images. Although we did not observe lesion hyperintensity on T1-weighted MR images due to fat content, use of fat-suppressed T1-weighted MR sequences or in-phase and out-of-phase T1-weighted MR sequences could be important to distinguish fat-containing tumors from those with sinusoidal dilatation, a feature frequently detected in our cases.
Strong hyperintensity on T2-weighted MR images is also a rare finding in FNH and has been reported in three of 42 lesions (7%) in a previous study (11). In that study, however, abnormal signal intensity on T2-weighted MR images was associated with a central scar in two lesions (11). In our series, lesion hyperintensity on T2-weighted MR images was a common finding, and in most cases, the hyperintensity was strong.
A central scar was present in 8% of our cases at imaging, and only one showed enhancement at delayed imaging and hyperintensity on T2-weighted MR images.
Persistent lesion enhancement on delayed-phase images was also a feature observed in 61% of the lesions in our study. This finding has been described only once in FNH, to our knowledge (16). Persistent contrast medium uptake in telangiectatic FNH could be related to sinusoidal dilatation. Therefore, none of the cases of telangiectatic FNH had all the patterns of typical FNH, including homogeneous lesions, isointensity or slight hyperintensity on T2-weighted MR images in relation to the surrounding liver parenchyma, strong arterial enhancement, and the presence of a scar.
In one of our patients, the diagnosis of telangiectatic FNH resulted from the occurrence of acute abdominal pain due to intralesional bleeding. Although no subcapsular or intraperitoneal hematoma developed, this finding, along with the association with oral contraceptives, makes these lesions particularly difficult to distinguish from adenoma on a clinical basis. We believe other studies of patients with unresected telangiectatic FNH are warranted to define the natural history of this lesion.
We were interested in the observation that 61% of the lesions in our study had delayed enhancement. Ichikawa et al (17) studied 25 patients with multiphasic helical CT with liver adenoma. Nine of their patients underwent delayed imaging, and only one lesion was hyperattenuating. This finding, along with the strong hyperintensity on T2-weighted MR images that we observed, can be a clue to the differential diagnosis with adenoma.
Our study has some limitations. First, the frequency of telangiectatic FNH in our study (11%) does not represent the real percentage of telangiectatic FNH among the general population with FNH, because our series included only surgical patients who had mainly atypical FNH. Second, since our study included patients in whom at least one lesion was resected because a certain diagnosis was not possible based on imaging criteria alone, we acknowledge that our series could present a selection bias. In the general population, telangiectatic FNH may show imaging features of typical FNH and may therefore be ignored. Third, we hypothesized that in patients with multiple lesions, all hypervascular lesions that were not resected were telangiectatic FNH; it is possible, however, that these lesions might have been adenomas, which are reported to coexist with FNH in 3.5%–6.0% of cases (4,9). Some of those lesions were biopsied intraoperatively, and a firm diagnosis of telangiectatic FNH was assigned on the basis of examination of the resected specimen. Other lesions were not sampled for biopsy, however, but in follow-up examinations, we observed no increase in lesion size or number.
In conclusion, telangiectatic FNH is an uncommon entity that differs from typical FNH at imaging. Lesions are multiple in 62% of cases. Atypical features of FNH often observed with telangiectatic FNH are lack of a central scar, lesion heterogeneity, hyperintensity on T1-weighted MR images, strong hyperintensity on T2-weighted MR images, and persistent contrast enhancement on delayed-phase contrast-enhanced CT or T1-weighted MR images. At least two of these features were observed in 44 of 61 lesions (72%).
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, P.A., V.V., G.B.; study concepts and design, V.V.; literature research, P.A., G.B.; clinical studies, P.A., V.V., B. Terris, B. Taouli, V.P., J.B.; data acquisition, P.A., V.V., G.B., B. Taouli; data analysis/interpretation, P.A., V.V., V.P., B. Terris, G.B.; statistical analysis, B. Taouli, G.B.; manuscript preparation, P.A., V.V., G.B.; manuscript definition of intellectual content, V.V., G.B.; manuscript editing and revision/review, V.V., G.B.; manuscript final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Kottke, M. Horger, H. Schimmel, and M. Wehrmann Atypical Focal Nodular Hyperplasia with Cluster-Like Internal Cysts Due to Fibrinoid Necrosis Am. J. Roentgenol., November 1, 2007; 189(5): W247 - W250. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lewin, A. Handra-Luca, L. Arrive, D. Wendum, V. Paradis, E. Bridel, J.-F. Flejou, J. Belghiti, J.-M. Tubiana, and V. Vilgrain Liver Adenomatosis: Classification of MR Imaging Features and Comparison with Pathologic Findings Radiology, November 1, 2006; 241(2): 433 - 440. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. D. Levy Invited Commentary RadioGraphics, January 1, 2004; 24(1): 18 - 19. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
