| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 Departments of Radiology (V.V., M.L., A.D., Y.M.)
2 Hepatology (D.V.)
3 Pathology (J.F.F.)
4 Digestive Surgery (J.B.), Hôpital Beaujon, 100 bd du General Leclerc, 92118 Clichy, France
5 Department of Digestive Surgery, Hôpital Antoine Beclere, Clamart, France (C.V.).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: The authors retrospectively studied images obtained in 23 patients with liver nodules who were being followed up for Budd-Chiari syndrome. Doppler ultrasonography was performed in all patients, computed tomography in 16, and magnetic resonance (MR) imaging in 20. The following lesion features were evaluated: location, number, size, vascularization, qualitative signal intensity at MR imaging, and homogeneity. Nodules were diagnosed on the basis of histopathologic findings or clinical and biologic data with no change at imaging during 2-year follow-up.
RESULTS: All patients had histopathologic features of chronic Budd-Chiari syndrome. Four patients had hepatocellular carcinoma (HCC), with one to three lesions. The mean diameter of the largest HCC lesion in each patient was 7.3 cm. All HCC lesions were heterogeneous and had high signal intensity on T2-weighted MR images. Nineteen patients had multiple benign regenerative nodules, most of which were smaller than 4 cm. Most nodules were homogeneous and hyperintense on T1- and T2-weighted images. In 15 patients, nodules were hypervascular in the arterial phase.
CONCLUSION: In patients with chronic Budd-Chiari syndrome, multiple (>10) small (<4-cm) lesions are suggestive of benignity.
Index terms: Hepatic veins, thrombosis, 761.659 • Liver, CT, 761.1211 • Liver, MR, 761.12143 • Liver, nodules, 761.3198 • Liver, US, 761.1298 • Liver neoplasms, blood supply • Liver neoplasms, diagnosis, 761.3198, 761.323
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The prognosis of patients with Budd-Chiari syndrome has been considered poor for many years. Recent advances in imaging techniques have enabled early diagnosis, recognition of asymptomatic cases, and surgical treatment (eg, portacaval anastomosis) and have resulted in better survival rates in these patients (2). At the same time, the nodules associated with Budd-Chiari syndrome have been described in the literature and include mostly hepatocellular carcinoma (HCC) and benign regenerative nodules (1,3–7), which are also called adenomatous hyperplastic nodules (4), nodular regenerative hyperplasia (5,6), and regenerative nodules (7). Very few studies have reported imaging findings about these nodules, and, to our knowledge, the vascularization of nodules in Budd-Chiari syndrome has not been evaluated with computed tomography (CT) or magnetic resonance (MR) imaging (4,5).
The objectives of this study were as follows: (a) to describe the imaging features of nodules in patients with Budd-Chiari syndrome, (b) to compare the imaging characteristics of benign nodules with those of HCC, and (c) to determine the etiology of these benign regenerative nodules.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Nodules were diagnosed on the basis of histopathologic analysis of specimens obtained by means of surgical resection or surgical or percutaneous biopsy or on the basis of clinical and biologic data combined with follow-up. Malignant cells at histologic examination, serum 
-fetoprotein level greater than 250 ng/mL, or both were the criteria for a malignant lesion. The absence of malignant cells with findings suggestive of regenerative nodules at histologic examination (8) or no change in the imaging appearance of the lesions for at least 2 years with serum -fetoprotein levels less than 10 ng/mL during follow-up were the criteria for a benign lesion.
The diagnosis of nodules was made at histopathologic examination in all but 10 patients. Two of the 10 patients had coagulation disorders, and eight were receiving anticoagulation therapy. Specimens were obtained after surgical resection in two patients, surgical biopsy in seven patients, and percutaneous needle biopsy in four patients. Histologic analysis of these specimens was reviewed by one experienced liver pathologist (J.F.F.). Microscopic analysis was performed, and the lesions were classified according to the guidelines of the International Working Party (8). The diagnosis of the nodules in the remaining 10 patients was made on the basis of clinical and biologic data combined with follow-up.
The images from the 23 patients with Budd-Chiari syndrome and associated hepatic nodules were retrospectively analyzed. There were five men and 18 women ranging in age from 22 to 64 (mean, 38.5) years. Nodules were discovered at diagnosis of hepatic venous outflow obstruction (n = 4) or during follow-up (n = 19). Follow-up ranged from 1 to 14 years after diagnosis of Budd-Chiari syndrome (mean, 6 years) (Fig 1). All patients had histopathologic features of chronic Budd-Chiari syndrome, including sinusoidal dilatation, centrolobular congestion, and areas of marked fibrosis. None of the patients had positive markers for viral hepatitis. The following etiologic factors were found in 20 patients: myeloproliferative syndrome (n = 12), which was associated in one patient with a membranous obstruction of the inferior vena cava; paroxysmal nocturnal hemoglobinuria (n = 3); coagulation disorders (n = 2); and antiphospholipid syndrome (n = 2). One patient had isolated membranous obstruction of the inferior vena cava. Eighteen patients had previously undergone creation of a portosystemic shunt. Shunts were mesocaval (n = 13), mesoatrial or innominate (n = 3), portacaval (n = 1), and mesorenal (n = 1).
|
MR imaging was performed with a 2.0-T (Elscint) (n = 18) or 0.5-T (MR Max Plus; GE Medical Systems) (n = 2) superconducting system. Conventional or fast spin-echo T2- and T1-weighted MR images were obtained in all but one patient, in whom only T1-weighted images were obtained. In addition, 19 patients received a bolus (0.1 mmol/kg) of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) before acquisition of dynamic gradient-recalled echo images.
Image Analysis
Image analysis of liver lesions was performed by two radiologists who were aware of the diagnosis of Budd-Chiari syndrome but not of the nodules. The diagnosis was made by consensus.
The following lesion criteria were evaluated: (a) location; (b) number; (c) size of the largest lesions in each patient; (d) sonographic pattern; (e) attenuation at unenhanced CT; (f) signal intensity at MR imaging; (g) lesion vascularization with contrast material–enhanced CT, MR imaging, or both; (h) homogeneous or heterogeneous appearance at US, CT, or MR imaging; and (i) presence of a peripheral rim at US, CT, or MR imaging. Three other items were also assessed: enlargement of the hepatic artery (defined as a main hepatic artery diameter larger than that of the splenic artery) on contrast-enhanced CT images or Doppler US scans, portal vein patency and direction at Doppler US, and patency of surgical portosystemic shunts on Doppler studies and/or contrast-enhanced CT and MR images, when available.
Follow-up
Twenty patients underwent follow-up for more than 2 years, including 19 patients with benign regenerative nodules and one patient with HCC. Comparison of the serial imaging features was made with US (20 patients), CT (eight patients), and MR imaging (12 patients). The mean follow-up was 3.6 years (range, 2–8 years).
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
-fetoprotein markers were normal in one patient and markedly increased in the three others (292, 1,600, and 10,200 ng/mL, respectively). A solitary lesion was found in three patients. These lesions were 4.5–8.0 cm in diameter (mean, 6.5 cm) (Fig 2). One patient had three lesions with diameters of 10.0, 2.0, and 1.5 cm. Therefore, the mean diameter of the largest lesion in each patient was 7.3 cm and that for all lesions was 5.5 cm. Four of the HCC lesions were located in the right lobe, and two were located in the left lobe. Lesion characteristics are summarized in Table 1. All tumors were heterogeneous on T2-weighted images. No peripheral rim was detected at US, CT, or MR imaging. Enlargement of the hepatic artery was observed in one patient.
|
|
|
Benign Regenerative Nodules
Nineteen patients had benign regenerative nodules, and all 19 had normal serum -fetoprotein markers (<10 ng/mL). In nine patients, the diagnosis was confirmed at histopathologic examination and the lesions were identified as benign multiacinar regenerative nodules (8). Nodules were composed of moderately enlarged hepatocyte plates (two to three cells thick) without nuclear atypia. In four patients, intranodular cholestasis was present, with an acinar architecture of hepatocytes around the bile plugs. When the specimen was large enough, portal tracts were identified within the nodule. It is interesting that although congestion was usually present in the extranodular parenchyma, it was absent in the nodules. In all but one patient, benign regenerative nodules were discovered during follow-up of Budd-Chiari syndrome from 1 to 14 years. Lesions were multiple in all patients. Fifteen of the 19 patients with benign regenerative nodules had more than 10 lesions within the liver at CT or MR imaging (Fig 3). Among these patients, 12 had more than 10 lesions at US and three patients had one lesion, two lesions, and three lesions, respectively. The other four patients had three to six lesions at US, CT, and/or MR imaging, and these lesions were disseminated throughout the liver in no specific location.
|
|
|
|
|
|
|
|
|
|
Analysis of follow-up images was possible in all patients with benign nodules, and only one patient with histopathologic diagnosis of nodule showed a change in the size and number of lesions. In this patient, the largest lesion enlarged from 3.0 to 7.5 cm during 3-year follow-up.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Benign Regenerative Nodules
Benign regenerative nodules have recently been described in the literature in association with Budd-Chiari syndrome (4–7). As in our series, the regenerative nodules were discovered only in patients with chronic Budd-Chiari syndrome, not in patients with acute disease. In some cases studied at imaging, authors showed that these lesions developed during follow-up (10–17 months after diagnosis) (4,12). We also showed that the lesions developed 1–14 years after diagnosis of Budd-Chiari syndrome in 18 of the 19 patients. An analysis of our patients and the literature did not reveal a specific cause for the development of these nodules. In this study and in the cases in the literature, only one patient with membranous obstruction of the vena cava had benign regenerative nodules. Despite the different terms that have been used to designate these benign regenerative nodules (ie, adenomatoid hyperplasia [4], nodular regenerative hyperplasia [5,6], or regenerative nodules [7]), these lesions are pathologically similar (5). No specific location has been identified, but Maia de Sousa et al (5) have suggested that microscopic nodules could be predominantly distributed in the periportal region. Because of the different sizes of these lesions, the term "nodular regenerative hyperplasia" is, as some authors have suggested, probably not appropriate and should be replaced with the term "multiacinar regenerative nodules" according to a recent classification proposed by the International Working Party (5,8). The two most characteristic findings concerning these nodules in the literature and in our series are the number and size of lesions. In most studies, the lesions were found to be numerous. In our study, 15 of 19 patients with benign nodules had more than 10 lesions on imaging studies. In livers removed at transplantation or autopsy, up to 50 nodules have been found (7). As noted in all published articles and most of the patients in the present series, nodule diameter can range from a few millimeters to 4.0 cm (4–6). In four of our patients, however, the diameter of the largest lesions ranged from 5.0 to 7 cm. In these four patients, more than 10 small nodules were also found in association with larger lesions.
Despite the number of previous studies about Budd-Chiari syndrome, imaging data for benign regenerative nodules in these reports are scanty, with only one report of MR imaging in two cases (4). In our experience, the appearance of benign regenerative nodules varies at US and is either hypo- or hyperechoic. These lesions are usually hyperattenuating in relation to the liver on unenhanced CT scans and strongly enhance at the arterial phase after the injection of contrast material. This hypervascularization was also found in two previous case reports with angiography, in which multiple hypervascular nodular areas were described. To our knowledge, however, it has not been described with CT or MR imaging (5,13). Soyer et al (4) showed hyperintense nodules on T1-weighted MR images and iso- or hyperintense nodules on T2-weighted MR images obtained in two patients. One of these patients had a hypointense rim at the periphery of a nodule. Similarly, in the present study, most lesions were hyperintense on T1-weighted MR images (12 of 16 patients [75%]) and hyper- or isointense on T2-weighted MR images (13 of 15 patients [87%]). Two patients had a nodule surrounded by a peripheral rim. Certain of these imaging features are atypical for benign regenerative nodules, such as hyperintensity on T1-weighted MR images, hyperintensity on T2-weighted MR images, and hypervascularization. To our knowledge, hyperintensity on T1-weighted MR images has not been reported in regenerative nodules but has been described in adenomatous hyperplastic nodules or "borderline" lesions in patients with cirrhosis (14). Although the reason for this hyperintense signal remains unclear, it is suggestive of premalignant lesions in patients with cirrhosis. As in other studies, we did not find fat deposits or peliosis in our hyperintense lesions on T1-weighted images. Copper accumulation, however, was not specifically studied.
Hyperintensity on T2-weighted images is not usually found in benign regenerative lesions and is observed in other lesions, including HCC, hepatocellular adenoma, and borderline lesions that contain foci of dysplasia or malignant transformations (14). Hypervascularization is also atypical in benign regenerative nodules. In patients with cirrhosis, Matsui et al (15) showed that the transformation from a predominantly portal blood supply to a predominantly arterial blood supply is suggestive of the malignant transformation of benign regenerative nodules. The findings of Matsui et al were confirmed with a morphologic study showing that angiogenesis occurs in the preneoplastic or early neoplastic stage during the course of hepatocarcinogenesis (16). Hypervascularization of regenerative nodules associated with Budd-Chiari syndrome probably does not have the same pathogenesis as that observed in HCC. In most of our cases, as in the study by Wanless (7), the caliber of the main hepatic artery was larger than that of the splenic artery, which suggests that an increase in arterial hepatic outflow might explain the hypervascularization of these lesions. Usually, the stimulus for liver regeneration is thought to be blood-borne, and prolonged exposure of hepatocytes to the action of hepatopoietins combined with other hepatotrophic factors in the portal blood flow is thought to increase hepatoproliferative activity (4,5). Only seven of our 19 patients with benign regenerative nodules had antegrade portal flow, whereas 12 had retrograde portal flow. This suggests that the cause of benign regenerative nodules is probably a disturbance in hepatic microcirculation secondary to obstruction of the hepatic vein.
The role of the respective changes in the arterial or portal circulation has not been clarified, but similar findings have been observed after congenital or surgical portacaval anastomosis (17–19). The first similarity is the development of hyperplastic foci that develop into true nodules. These nodules are large, numerous, and may measure up to 1.0 cm in diameter within 6 months after surgical anastomosis (17). Furthermore, at first they involve predominantly periportal cells (17). In an experimental study of portacaval anastomosis in rats, nodules were detected in all animals 6–12 months after the procedure and resulted in markedly irregular nodularity after 18 months (17). The second similarity is abnormal portal blood flow such as a decrease in portal blood flow secondary to portacaval anastomosis or total deprivation of portal blood in congenital portacaval shunts (20), and the third is an increase in hepatic arterial inflow resulting in the development of multiple arteries with markedly enlarged lumens and thickened walls within the portal triads in congenital and surgical portacaval shunts (19,20).
In patients with Budd-Chiari syndrome, the relationship between the development of nodules and portacaval anastomosis may be hypothesized; however, of the 19 patients in this study with benign nodules, three had no shunts and five had complete obstruction of the surgical portacaval anastomosis.
The prevalence of benign regenerative nodules associated with Budd-Chiari syndrome is not known. On the basis of the literature, the prevalence appears to be low; however, many lesions were probably undetected. In a pathologic study of 10 livers removed at transplantation or autopsy (7), all 10 had small, 1.0-mm-diameter nodules and eight to 10 had larger nodules. Furthermore, in our clinical study of 77 patients with Budd-Chiari syndrome, benign regenerative lesions were detected with imaging techniques in 25%. The true prevalence of benign regenerative lesions, however, is not known, and the fact that a large number of our patients had portosystemic shunts could perhaps increase erroneously the real prevalence.
Our study has several limitations. First, the diagnosis was confirmed histopathologically in only half of the patients with benign regenerative nodules. Furthermore, in the patients with benign regenerative nodules proved by means of histopathologic analysis, specimens were obtained only from the largest nodules; the diagnosis of the other lesions was not confirmed with histopathologic examination. In the other patients, however, the lesions were considered to be benign and only patients without change during a follow-up of at least 2 years were included. Absence of change during 2-year follow-up cannot exclude rare forms of slowly growing HCC; however, these lesions are very occasionally encountered. Second, the evolution of benign regenerative nodules cannot be confirmed even change was not observed in 18 of 19 patients after 2–8-year follow-up. Third, this study is retrospective and imaging techniques such as CT or MR imaging were not available to all patients. In most patients, however, lesion vascularization was assessed with either CT or MR imaging.
In conclusion, in our experience, chronic Budd-Chiari syndrome is more often associated with benign regenerative nodules than with HCC. The distinction between these two lesions cannot be made on the basis of signal intensity at MR imaging or lesion hypervascularization. On the basis of our experience and that reported in the literature, benign nodules can be diagnosed with reasonable confidence in patients with more than 10 nodules smaller than 4 cm in diameter at imaging. The hyperattenuation of nodules at unenhanced CT could be another tool for differentiating regenerating nodules from HCC because, in our study, 61% of the regenerating nodules and none of the HCCs were hyperattenuating. In these cases, however, CT or MR imaging follow- up is necessary.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
From the 1997 RSNA scientific assembly.
Abbreviation: HCC = hepatocellular carcinoma
Author contributions: Guarantor of integrity of entire study, V.V.; study concepts and design, V.V.; definition of intellectual content, V.V., M.L.; literature research, M.L.; clinical studies, C.V., D.V., J.F.F.; experimental studies, J.B.; data acquisition, A.D., M.L.; data analysis, V.V., M.L.; manuscript preparation and editing, V.V., M.L.; manuscript review, V.V., M.L., A.D., Y.M.
Received December 5, 1997;
revision requested February 24, 1998; revision received July 17, 1998;
accepted September 23, 1998.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R Moucari, P-E Rautou, D Cazals-Hatem, A Geara, C Bureau, Y Consigny, C Francoz, M-H Denninger, V Vilgrain, J Belghiti, et al. Hepatocellular carcinoma in Budd-Chiari syndrome: characteristics and risk factors Gut, June 1, 2008; 57(6): 828 - 835. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Willatt, H. K. Hussain, S. Adusumilli, and J. A. Marrero MR Imaging of Hepatocellular Carcinoma in the Cirrhotic Liver: Challenges and Controversies Radiology, May 1, 2008; 247(2): 311 - 330. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Brancatelli, V. Vilgrain, M. P. Federle, A. Hakime, R. Lagalla, R. Iannaccone, and D. Valla Budd-Chiari Syndrome: Spectrum of Imaging Findings Am. J. Roentgenol., February 1, 2007; 188(2): W168 - W176. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Rikitake, T. Itoh, M. Natsuaki, Y. Katayama, S. Ohtsubo, and H. Kataoka Surgical Repair for Entire Inferior Vena Cava Occlusion in Budd-Chiari Syndrome Asian Cardiovasc Thorac Ann, October 1, 2006; 14(5): e88 - e90. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Bargallo, R. Gilabert, C. Nicolau, J. C. Garcia-Pagan, J. R. Ayuso, and C. Bru Sonography of Budd-Chiari syndrome. Am. J. Roentgenol., July 1, 2006; 187(1): W33 - W41. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Zeitoun, G. Brancatelli, M. Colombat, M. P. Federle, D. Valla, T. Wu, C. Degott, and V. Vilgrain Congenital Hepatic Fibrosis: CT Findings in 18 Adults Radiology, April 1, 2004; 231(1): 109 - 116. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. K. Hussain, I. Syed, H. V. Nghiem, T. D. Johnson, R. C. Carlos, W. J. Weadock, and I. R. Francis T2-weighted MR Imaging in the Assessment of Cirrhotic Liver Radiology, March 1, 2004; 230(3): 637 - 644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Krinsky and G. Israel Nondysplastic Nodules That Are Hyperintense on T1-Weighted Gradient-Echo MR Imaging: Frequency in Cirrhotic Patients Undergoing Transplantation Am. J. Roentgenol., April 1, 2003; 180(4): 1023 - 1027. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Brancatelli, M. P. Federle, L. Grazioli, R. Golfieri, and R. Lencioni Benign Regenerative Nodules in Budd-Chiari Syndrome and Other Vascular Disorders of the Liver: Radiologic-Pathologic and Clinical Correlation RadioGraphics, July 1, 2002; 22(4): 847 - 862. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Maetani, K. Itoh, H. Egawa, H. Haga, T. Sakurai, N. Nishida, F. Ametani, T. Shibata, T. Kubo, K. Tanaka, et al. Benign Hepatic Nodules in Budd-Chiari Syndrome: Radiologic--Pathologic Correlation with Emphasis on the Central Scar Am. J. Roentgenol., April 1, 2002; 178(4): 869 - 875. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Brancatelli, M. P. Federle, L. Grazioli, R. Golfieri, and R. Lencioni Large Regenerative Nodules in Budd-Chiari Syndrome and Other Vascular Disorders of the Liver: CT and MR Imaging Findings with Clinicopathologic Correlation Am. J. Roentgenol., April 1, 2002; 178(4): 877 - 883. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Baron and M. S. Peterson From the RSNA Refresher Courses: Screening the Cirrhotic Liver for Hepatocellular Carcinoma with CT and MR Imaging: Opportunities and Pitfalls RadioGraphics, October 1, 2001; 21(90001): S117 - 132. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Krinsky, V. S. Lee, N. D. Theise, J. C. Weinreb, N. M. Rofsky, T. Diflo, and L. W. Teperman Hepatocellular Carcinoma and Dysplastic Nodules in Patients with Cirrhosis: Prospective Diagnosis with MR Imaging and Explantation Correlation Radiology, May 1, 2001; 219(2): 445 - 454. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
2 mm) shows the high-signal-intensity lesion (arrowheads). Ascites and dilatation of the azygos (A) and hemiazygos (H) veins are seen.
