- Email: [email protected]
Congenital portosystemic shunts associated with liver tumours
Clinical Radiology 68 (2013) e362ee369
Contents lists available at SciVerse ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
Congenital portosystemic shunts associated with liver tumours L.F. Pupulim a, *, M.-P. Vullierme a, V. Paradis b, D. Valla c, S. Terraz d, V. Vilgrain a ˇ
ˇ
Department of Radiology, Hopital Beaujon, Assistance Publique des Hopitaux de Paris, Universit e Paris Diderot, Sorbonne Paris Cit e, INSERM unit e 773 CRB3, Paris, France b Department of Pathology, Hopital Beaujon, Assistance Publique des Hopitaux de Paris, Universit e Paris Diderot, Sorbonne Paris Cit e, INSERM unite 773 CRB3, Paris, France c Department of Hepatology, Hopital Beaujon, Assistance Publique des Hopitaux de Paris, Universit e Paris Diderot, Sorbonne Paris Cit e, INSERM unit e 773 CRB3, Paris, France d Department of Radiology, University Hospitals of Geneva, Geneva, Switzerland ˇ
a
ˇ
ˇ
ˇ
article in formation
AIM: To evaluate the diagnosis and presentation of liver tumours in patients with congenital portosystemic shunts (CPS). MATERIALS AND METHODS: Eight patients were diagnosed in Hopital Beaujon as having CPS. All patients underwent Doppler ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), and histological examination of liver tumours. CPS were classified according to anatomy and the amount of portal flow deviated to the systemic circulation as: total, subtotal, or partial. Liver tumours were diagnosed by needle core biopsy (n ¼ 5) or surgery (n ¼ 3). Clinical follow-up was available in all patients but one (mean follow-up 36 months; range 1e5 years). RESULTS: Six patients had total CPS, one patient had a subtotal CPS, and the last had a partial CPS. All patients presented with multiple liver nodules (range four to >15). The tumours were characterized as focal nodular hyperplasia (FNH; n ¼ 4), FNH with hepatocellular adenoma (n ¼ 2), and regenerative nodular hyperplasia (n ¼ 2). In four of seven patients (57%) that had follow-up, tumours showed enlargement or new lesions appeared. CONCLUSION: In this series of CPS patients, tumours were all benign, multiple, and of hepatocellular origin, and different tumours were present simultaneously in two patients. Tumour enlargement or new nodules were common during follow-up. Ó 2013 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. ˇ
Article history: Received 29 August 2012 Received in revised form 16 January 2013 Accepted 22 January 2013
Introduction Portosystemic shunts in patients without a history of trauma, liver biopsy, portal thrombosis, cirrhosis, or portal
* Guarantor and correspondent: L.F. Pupulim, Department of Radiology, Paris Hopital Beaujon, Assistance Publique des Hopitaux de Paris, Universite , INSERM unite 773 CRB3, 100 Blvd du Ge ne ral Diderot, Sorbonne Paris Cite Leclerc, 92110 Clichy, France. Tel.: þ33 (0) 1 40 87 53 55. E-mail address: [email protected] (L.F. Pupulim).
hypertension are considered to be congenital in origin. The latter shunts result from abnormalities in the complex embryonic development of the hepatic venous system. Recently, the extensive use of advanced imaging methods has increased the diagnosis of congenital portosystemic shunts (CPSs) and a wide variety of connections has been reported in medical literature, often as case reports. CPSs are frequently reported in association with other congenital malformations, which strengthen the embryological aetiology of such shunts. Despite being present early in the life, CPSs may remain asymptomatic until adolescence or
ˇ
ˇ
0009-9260/$ e see front matter Ó 2013 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.crad.2013.01.024
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
e363
Figure 1 A 38-year-old woman with total extra-hepatic CPS associated with multiple liver nodules diagnosed as RNH at biopsy (patient 5). (a) Axial cross-sectional T2-weighted MRI image showing the total extra-hepatic CPS. Splenic and superior mesenteric veins form a trunk that turns in the direction of the left renal vein (*). (b) CT imaging during the portal venous phase at the same level as (a) shows four ill-defined, hypo-attenuating liver nodules (black arrows). In this case, liver nodules were more conspicuous at CT than MRI.
adulthood.1,2 The morphological presentation of these shunts can show enormous variability: they can be intra- or extra-hepatic, simple or multiple, large or small, and even aneurysmal. Current classifications are based on the location of the CPS (intra- or extra-hepatic shunts) and are not related to degree of hepatic portal blood deprivation.3,4 CPSs are known to be associated with the development of
liver tumours, such as focal nodular hyperplasia (FNH),5e9 regenerative nodular hyperplasia (RNH),10,11 and hepatocellular adenoma.12,13 These tumours are considered to result from an abnormal hepatocellular response to impairment of the portal venous supply and compensatory increases in hepatic arterial flow.14e17 In the present study, the medical reports and imaging findings of eight
Figure 2 Subtotal intra-hepatic CPS in a 25-year-old woman presenting with numerous small FNH nodules and a large nodule in segment 6 diagnosed as FNH with foci of adenoma after orthotopic liver transplantation (patient 4). (a) Portography after injection of iodine contrast medium in the superior mesenteric artery shows a portosystemic shunt between the main portal vein and the atrium (black arrow). An atrophic portal branch is also seen (white arrow) defining this shunt as subtotal. (b) T1-weighted MRI image showing a hyperintense large tumour in segment 6 (white arrow). (c) Arterial-phase CT image shows that this tumour has increased in size at the follow-up 2 years later (white arrow). This was diagnosed as an FNH with internal foci of adenoma on the histopathological analysis after liver transplantation. Note also the enlarged hepatic artery (black arrow).
e364
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
Figure 3 Partial intra-hepatic CPS in an asymptomatic 23-year-old patient (patient 7). (a) Maximum intensity projection reconstructed CT image of the portal phase showing a large portosystemic shunt (*) connecting the portal trunk to the IVC. Other portal branches are present. LPV, left portal vein; RAPV, right anterior portal vein. This image also shows small calcifications that were present in an iso-attenuating FNH (black arrows) diagnosed on biopsy. (b) Cross-sectional, contrast-enhanced, axial CT image at the portal-phase at the level of the LPV. A small hypovascular non-biopsied nodule (small black arrow) and a large hypervascular nodule are seen (large black arrow). The latter was confirmed as a FNH nodule on histopathological analysis of a liver biopsy. The CPS is also seen in transverse section (*).
Table 1 Clinical setting, embryological anomalies, histopathological findings, and imaging features of liver tumours of eight patients presenting congenital portosystemic shunt (CPS). Patient
Sex; age at diagnosis
Clinical context at presentation
Laboratory findingsa
Type of portosystemic shunt
Vascular anomalies
Associated congenital anomalies
Follow-up
1
F; 24
Abdominal pain
GGT (5N); AP (2N); AST/ALT (1,5 N)
Total, extrahepatic, single
Trunk SMV þ SV to IVC
No
2
F; 14
No symptoms
GGT (4N); AP (5N); AST/ALT (3N)
Total, extrahepatic, single
Trunk SMV þ SV to right atrium
3
M; 13
Right upper quadrant pain
GGT (3N); AP (2N); bilirubin (1,2N); AST/ALT (1,5 N).
Total, intrahepatic, single
RPV to RHV through an aneurysm. No other portal branches
Ataxiatelangiectasia syndrome; intestinal malrotation Turner syndrome; aorta coarctation No
4
F; 25
No symptoms
GGT (2N)
Subtotal, intrahepatic, single
LPV to right atrium through a PDV. No right portal vein
No
5
F; 38
Abdominal pain
Normal values
F; 16
Right upper quadrant pain
7
F; 23
Abdominal pain
GGT (4N); AP (2N); serum ammonia (1,5N) Normal values
Trunk SMV þ SV to left renal vein Large tube connecting SV to left renal vein
No
6
Total, extrahepatic, single Total, extrahepatic, single Partial, intrahepatic, single
RPV to IVC in the confluence of RHV
8
F; 24
No symptoms
Normal values
Total, intrahepatic, single
RPV to IVC without other portal branches
Lobar pulmonary sequestration No
No
60 months e Tumour enlargement from 7 to 13 cm 36 months e Surgery for tumours that presented enlargement 27 months e Surgery for tumours that presented enlargement 41 months e no changes 59 months e no changes 9 months - multiples small enhancing nodules appeared 24 months e no changes
F, female; M, male; AP, serum alkaline phosphatase; GGT, serum gamma-glutamyl-transferase activity; AST/ALT: transaminases; SMV, superior mesenteric vein; SV, splenic vein; IVC, inferior vena cava; RPV, right portal vein; RHV, right hepatic vein; LPV, left portal vein; PDV, persistent ductus venosus; OLT, orthotopic liver transplantation, US, ultrasound; NE, non-enhanced; CE, contrast enhanced; CT, computed tomography; MRI, magnetic resonance imaging; NCB, needle core biopsy; FNH, focal nodular hyperplasia; RNH, regenerative nodular hyperplasia; IV, intravenous; nod, nodule. a Only abnormal results are mentioned in laboratory findings. Numbers in parentheses express the times serum levels were above upper normal values (N). b After OLT the pathological liver analysis showed multiples FNH not identified on imaging.
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
consecutive patients with CPS were retrospectively analysed with the aim of evaluating the diagnosis and presentation of hepatic tumours in patients with CPS.
Material and methods During a 15 year period (1993e2008) eight patients were diagnosed in Hopital Beaujon with CPS associated with liver tumours. The medical data, imaging findings, and histopathology findings were retrospectively reviewed. There was no history of abdominal surgery, trauma, or liver biopsy before the diagnosis of the shunt. Cirrhosis, chronic hepatitis, portal vein thrombosis, and portal hypertension were ruled out in all patients following assessment of the imaging, clinical, and laboratory findings. Serum alpha-fetoprotein and carcinoembryonic antigen were normal in all patients. Clinical follow-up was available in all patients but one. The mean total follow-up was 36 months (range 1e5 years). ˇ
Imaging technique Doppler ultrasound, abdominal multiphasic computed tomography (CT), and magnetic resonance (MR) imaging were performed in all patients. Five patients also underwent
e365
selective hepatic angiography of liver vessels. Angiography was not performed in the other three patients because of vascular reconstructions available on CT and/or MRI that were considered satisfactory to evaluate hepatic vascular anatomy. The present study was a retrospective analysis of examinations performed throughout a period of 15 years; as a result, protocols of the CT and MRI examinations were not uniform for all eight patients. Different CT machines were used (CT Twin Flash; Marconi Medical Systems, Haifa, Israel; CT Lightspeed with 8 and 64 detectors, GE Healthcare, Milwaukee, WI, USA). Non-ionic intravenous contrast material (120e160 ml; 350 mg iodine/ml) was administered using an automatic power injector. Arterial and portal venous acquisitions were performed. MRI was performed using two different 1.5 T units (Elscint, Haifa, Israel) and (Intera, Philips Medical Systems, Best, The Netherlands) with acquisition of T1 and T2-weighted images in gradient echo (GE) and/or fast spin-echo techniques. Contrast-enhanced T1-weighted GE images were obtained after intravenous injection of 0.1 mmol/kg of body weight of non-specific chelates of gadolinium. Digital hepatic angiography (Multistar T.O.P.; Siemens; Munich, Germany) examinations were performed by catheterization of the coeliac trunk using the Seldinger technique. Arterial and venous phases were performed, the
Outcome
Number of nodules on imaging
Nodules size range (diameter in cm)
Pathologic proof
Tumour diagnosis
NE-CT (attenuation)
MRI T1 (intensity)
MRI T2 (intensity)
Enhancement on arterial phase of IV contrast
No treatment performed
6 (3 large and 3 small)
1e10
NCB (2 largest nodules)
FNH
Hypo, heterogeneous
Iso (large nodules: centre hypo)
Heterogeneous; hyper in centre (large nodules)
Yes
Clinically stable
>15
1e13
Surgical biopsy (3 nodules)
FNH
Iso (largest nodule: hypo)
Hyper
Hypo
Yes
2e15
Surgery (OLT)
Hypo
Hyper (2 nod); Hypo (2 nod)
No
1e12
Surgery (OLT)
Hypo
Hyper (2 nod); Hypo (2 nod) Hyper
Iso; slightly hyper
Yes (heterogeneous)
Hypo, iso
Iso
Hyper, iso
No
4b
Clinically stable
>15
1e5
NCB (1 nod)
One nodule: adenoma Other nodules: FNH. FNH with foci of adenoma (large nodules). Small nodules: FNH. RNH
Clinically stable
>15
1e4
NCB (2 nod)
RNH
Hypo, iso
Hyper
Hyper
Yes
Clinically stable
>10
1e7
NCB (2 nod)
FNH
Hypo
Iso
Iso
Yes
Clinically stable
6
1e4
NCB (1 nod)
FNH
Hypo
Hyper
Iso; Hyper (largest)
Yes (2 largest); No (others)
OLT for severe encephalopathy after tumour resections OLT for severe encephalopathy after tumour resections
>8
e366
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
latter indirectly evaluated the portal vein and the portosystemic shunts after injection of iodine-containing contrast medium in the superior mesenteric and/or splenic artery.
Imaging analysis Images were retrospectively reviewed by two experts in hepatic imaging in consensus and CPSs were defined according to the degree of hepatic portal blood deprivation as follows: (a) total: when all portal blood flow was seen to be deviated to the systemic circulation on imaging and no intrahepatic portal branches could be identified (Fig 1); (b) subtotal: most of the portal blood flow was deviated to the systemic circulation but some abnormal atrophic intra-hepatic portal branches could be identified on imaging (Fig 2); (c) partial: besides the shunt, normal intra-hepatic portal branches could be identified, meaning that some of the portal blood still passed through the liver sinusoids (Fig 3). Moreover, according to the degree of portal blood deprivation, total and subtotal CPSs were considered as major. CPSs were also defined as intra-hepatic when the abnormal connection was located within the liver and as extra-hepatic when the shunt had no relation to the liver. The number of shunt pathways was also assessed to define them as single or multiple. Liver tumours were analysed according to: (a) number of tumours, (b) size, (c) attenuation on unenhanced CT, (d) lesion intensity on T1- and T2-weighted MRI sequences, and (e) enhancement on contrast-enhanced studies (Table 1). Typical imaging findings for FNH,15 hepatocellular adenoma,18,19 and RNH17 were analysed. For FNH, imaging findings included: lesion homogeneity; intense and homogeneous contrast enhancement on arterial-phase imaging with lack of wash-out; central scar; central arteries on Doppler ultrasound; and absence of capsule. For telangiectatic/inflammatory hepatocellular adenomas, imaging findings included: marked hypersignal on T2-weighted sequences; strong arterial enhancement with persistent enhancement in portal and delayed phases; and absence of signal dropout on chemical shift sequences. For steatotic hepatocellular adenomas, imaging findings included: signal dropout on T1-weighted chemical shift sequence; isosignal or slight hypersignal on T2-weighted images; and moderate enhancement in the arterial phase with no persistent enhancement in the portal and delayed phases. The diagnosis of RNH was suspected when a lesion was of high signal intensity on T1-weighted images and iso- or slightly hypersignal intensity on T2-weighted images. Tumour enlargement and the appearance of new tumours were determined during follow-up. The hepatic artery was also evaluated, and it was considered dilated when the diameter at the origin was greater than that of the splenic artery.
Histopathological analysis Needle core biopsy (NCB) was performed in all patients sampling either one or two nodules (Table 1). As patients presented with multinodular liver disease, target nodules were chosen according to the possibilities of best access. Biopsy of the adjacent liver was also obtained in all patients.
One patient had a corresponding surgical biopsy where three nodules were sampled. In two patients, additional tumour cell clonal tests were performed and further histopathological analysis was obtained after orthotopic liver transplantation (OLT). Histopathological records of liver tumours and liver parenchyma were retrospectively reviewed by an experienced liver pathologist. Tumour nodules were reviewed after haematoxylin and eosin (H&E), trichrome, and reticulin staining. FNH was considered when the lesion was well-circumscribed and composed of benign-appearing hepatocytes arranged in nodules that were partly or completely delineated by fibrous septa. Dystrophic vessels were observed inside the fibrous septa, focally associated with ductular proliferation. The hepatocytes were hyperplastic, arranged in liver plates of normal or slightly increased thickness with a well-preserved reticulin framework. When the hepatocellular component appeared more proliferative and extensive, a diagnosis of FNH with adenomatous features was considered. RNH was defined as nodular architecture of the liver parenchyma without extensive fibrosis, showing areas of thickened liver cell plates associated with areas of atrophic cell plates. Liver samples without tumour were histologically assessed after H&E and trichrome staining. The extent of portal fibrosis, presence of sinusoidal dilatation, necroticoinflammatory changes, and intensity of steatosis were systematically evaluated.
Results Clinical findings The patients included seven women and one man with a mean age of 22-years-old (range 13e38 years). Three patients had incidental discovery of CPSs and five presented with abdominal pain. An associated congenital anomaly was found in three patients as follows: ataxia-telangiectasia syndrome and intestinal malrotation (n ¼ 1); Turner’s syndrome and coarctation of the aorta (n ¼ 1), and lobar pulmonary sequestration (n ¼ 1). A summary of the clinical findings, liver blood tests, histopathological findings, and imaging features are shown in Table 1.
CPSs A CPS was seen during Doppler ultrasound in all patients. Contrast-enhanced CT confirmed the presence of portosystemic shunts in all patients. Six patients had a total CPS; four were extra-hepatic and these shunts were connected to the inferior vena cava (IVC; n ¼ 1), the left renal vein (n ¼ 2), or the right atrium (n ¼ 1). Among the two total intra-hepatic portosystemic shunts, one presented as an aneurysm connection between the right portal vein and the right hepatic vein, and the other connected the right portal vein to the IVC. One patient had a subtotal CPS with an atrophic intra-hepatic portal branch (Fig 2). One patient had a partial CPS between the right branch of the portal vein and the IVC (Fig 3). The intra-hepatic portal branches were normal in this patient. All shunts had a single pathway
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
defining an individual shunt. The hepatic artery was dilated in all patients with total and subtotal shunts (n ¼ 7) with a diameter ranging from 7e9 mm. It was considered of normal calibre in the patient with a partial shunt.
Liver nodules All eight patients presented with multiple liver tumours ranging from four to greater than 15 nodules. Four patients had FNH, two had RNH, one (patient 3) presented multiple FNH and a single hepatocellular adenoma, and another patient (patient 4) was diagnosed with multiple FNH nodules and three of the largest FNH nodules were associated with foci of hepatocellular adenoma. Imaging features of liver tumours are summarized in Table 1. Among the four patients with FNH, only two (patients 1 and 8) had all imaging features of typical FNH15 (Fig 4). The liver adenoma and the FNH nodules with foci of liver adenoma were hypo-attenuating on non-enhanced CT images and presented with heterogeneous enhancement on arterial-phase imaging. The nodules in these patients (patients 3 and 4) did not show typical findings of FNH either of telangiectatic/inflammatory or steatotic hepatocellular adenoma. These two patients underwent tumour resection because of a preoperative diagnosis of hepatocellular adenoma on percutaneous biopsy and tumours enlarged during
e367
follow-up. After surgery they experienced a degree of hepatic encephalopathy and persistent hepatic failure and finally underwent liver transplantation. Typical imaging features of RNH were found in one of two patients. One patient was lost to follow-up. Three patients presented with nodule enlargement during followup. One patient (patient 7), with four nodules at the time of diagnosis, developed multiple small hypervascular liver nodules during the follow-up. The tumours remained stable in the other three patients.
Discussion The current series of patients with CPS analyses both the variety of liver tumours associated with CPS as well as the morphological aspects of these interesting congenital shunts. Liver tumours, such as FNH, are known to develop in response to focal or diffuse abnormal perfusion of liver.5e17,20e24 In the normal liver, they are usually solitary, whereas they are usually multiple in patients with vascular disorders such as BuddeChiari syndrome, congenital hepatic fibrosis, OslereRendueWeber disease, and cavernous transformation of the portal vein. Similarly, FNH have been described in association with CPSs.5e9 FNH were by far the most common tumours observed in this series. CPS results in
Figure 4 Total intrahepatic CPS in a 24-year-old woman (patient 8). (a) Arterial-phase T1-weighted MRI image showing two enhanced liver nodules (black arrows). CPS is seen in transverse section (*). (b) Portal-phase, T1-weighted MRI image (more cranial section) shows a typical enhanced FNH of the liver dome (black arrow) with a lobulate contour and central scar. Diagnosis was confirmed by biopsy. CPS (*) is almost joining the IVC at this level. (c) T2-weighted MRI image of the same nodule as (b) showing a hyperintense central scar within the FNH (black arrow).
e368
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
portal venous deprivation, which is often complete, and this induces a compensatory increase in arterial inflow. Indeed, arterial compensation was confirmed in all cases with major portal deprivation (either total or subtotal CPS). The only patient who did not have an enlarged hepatic artery had a partial shunt, suggesting that haemodynamic changes are more prominent in major shunts than in partial ones. The present series also confirms the possible association of other hepatocellular lesions with CPS, such as hepatocellular adenoma and RNH. Two of the present patients (25%) presented simultaneously with FNH and hepatocellular adenoma. The diagnosis in the case of FNH with foci of hepatocellular adenoma was confirmed by immunohistochemical analysis and tumour cells clonal pattern. The mechanisms that lead to the development of monoclonal tumours within polyclonal nodules have to be elucidated. Hence, CPS may play a role in such cases. In view of the possibility of malignant transformation of hepatocellular adenoma, patients with CPS should be closely followed up. In the present series, all patients were adults and had multiple liver tumours. Hepatologists and radiologists should be alerted to the possibility of CPS in young adults presenting with spontaneous multiple liver tumours. Imaging, and especially MRI, plays an important role in diagnosing FNH. The reported sensitivities and specificities are 70% and 98%, respectively.15 In the present series of patients with CPS, FNHs were atypical in 50% of patients (two out of four). It is possible that complete and chronic portal deprivation could alter the morphology of FNH. The variety of lesions that are encountered in patients with CPS should reinforce the role of liver biopsy, notably in patients with atypical lesions at imaging. In four of seven patients with serial imaging, liver tumours enlarged in size or new nodules appeared. To the authors’ knowledge, this is the first such description and suggests that CPS may be a progressive disease. Follow-up and treatment, as well as CPS closure, may be considered for these patients.25 Excess involution and abnormal connections of the periintestinal vitelline venous network are involved in the origin of CPS.26e28 The diversity of CPS in terms of size, localization, number of pathways, and possible association with normal portal intra-hepatic anatomy is related to the complex development of the embryonic hepatic venous system. Previous classifications based on morphology were proposed and separate shunts according to their level, either intra-hepatic3 or extra-hepatic.4 However, these methods fail to classify all cases and do not evaluate the degree of deprivation of portal blood flow. The morphology of CPS could be described according to the presence and degree of intra-hepatic portal venous flow as highlighted here. This classification may better explain the relationship between CPS and hepatocellular lesions. Most of the present patients with hepatocellular lesions had a major shunt (either total or subtotal). One limitation of the present study was the small number of patients, although all clinical and morphological details were available. Although CPS is rare, further studies with larger number of patients are necessary to evaluate progressive changes in liver tumours in patients with CPS.
In conclusion, although CPS is a congenital disease, in the present study a number of patients were diagnosed in adulthood with a number of benign hepatocellular tumours. These lesions were all benign and multiple. Varying types of tumour (FNH and hepatocellular adenoma) may coexist in the same patient. During follow-up, tumour enlargement or new nodules appeared in over half of the present patients.
Acknowledgements The authors thank the substantial assistance of Dr Dermot O’Toole (Department of Clinical Medicine & Gastroenterology of St James’s Hospital & Trinity College Dublin, Ireland) for editing the manuscript.
References 1. Kakitsubata Y, Kakitsubata S, Kiyomizu H, et al. Intrahepatic portalhepatic venous shunts demonstrated by US, CT and MR imaging. Acta Radiol 1996;37:680e4. 2. Murray CP, Yoo SJ, Babyn P. Congenital extrahepatic portosystemic shunts. Pediatr Radiol 2003;33:614e20. 3. Park JH, Cha SH, Han JK, et al. Intrahepatic portosystemic venous shunt. AJR Am J Roentgenol 1990;155:527e8. 4. Morgan G, Superina R. Congenital absence of the portal vein: two cases and a proposed classification system for portosystemic vascular anomalies. J Pediatr Surg 1994;29:1239e41. 5. Altavilla G, Cusatelli P. Ultrastructural analysis of the liver with portal vein agenesis: a case report. Ultrastruct Pathol 1998;22:477e83. 6. Guariso G, Fiorio S, Altavilla G, et al. Congenital absence of the portal vein associated with focal nodular hyperplasia of the liver and cystic dysplasia of the kidney. Eur J Pediatr 1998;157:287e90. 7. Tanaka Y, Takayanagi M, Shiratori Y, et al. Congenital absence of portal vein with multiple hyperplastic nodular lesions in the liver. J Gastroenterol 2003;38:288e94. 8. Kim T, Murakami T, Sugihara E, et al. Hepatic nodular lesions associated with abnormal development of the portal vein. AJR Am J Roentgenol 2004;183:1333e8. 9. De Gaetano AM, Gui B, Macis G, et al. Congenital absence of the portal vein associated with focal nodular hyperplasia in the liver in an adult woman: imaging and review of the literature. Abdom Imaging 2004;29:455e9. 10. Grazioli L, Alberti D, Olivetti L, et al. Congenital absence of portal vein with nodular regenerative hyperplasia of the liver. Eur Radiol 2000;10:820e5. 11. Arana E, Martí-Bonmatí L, Martínez V, et al. Portal vein absence and nodular regenerative hyperplasia of the liver with giant inferior mesenteric vein. Abdom Imaging 1997;22:506e8. 12. Nakasaki H, Tanaka Y, Ohta M, et al. Congenital absence of the portal vein. Ann Surg 1989;210:190e3. 13. Kamiya S, Taniguchi I, Yamamoto T, et al. Analysis of intestinal flora of a patient with congenital absence of the portal vein. FEMS Immunol Med Microbiol 1993;7:73e80. 14. Wanless IR, Lentz JS, Roberts EA. Partial nodular transformation of liver in an adult with persistent ductus venosus. Arch Pathol Lab Med 1985;109:427e32. 15. Hussain SM, Terkivatan T, Zondervan PE, et al. Focal nodular hyperplasia: findings at state-of-the-art MR imaging, US, CT, and pathologic analysis. RadioGraphics 2004;24:3e19. 16. Wanless IR, Mawdesley C, Adams R. On the pathogenesis of focal nodular hyperplasia of liver. Hepatology 1985;5:1194e200. n M, Pe rez A, et al. Clinical and radiological 17. Alonso-Gamarra E, Parro manifestations of congenital extrahepatic portosystemic shunts: a comprehensive review. RadioGraphics 2011;31:707e22. 18. Laumonier H, Bioulac-Sage P, Laurent C, et al. Hepatocellular adenomas: magnetic resonance imaging features as a function of molecular pathological classification. Hepatology 2008;48:808e18.
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369 19. Ronot M, Bahrami S, Calderaro J, et al. Hepatocellular adenomas: accuracy of magnetic resonance imaging and liver biopsy in subtype classification. Hepatology 2011;53:1182e91. 20. Wanless IR. Micronodular transformation (nodular regenerative hyperplasia) of the liver: a report of 64 cases among 2500 autopsies and a new classification of benign hepatic nodules. Hepatology 1990;11:787e97. 21. Weinbren K, Washington SLA. Hyperplastic nodules after portacaval anastomosis in rats. Nature 1976;264:440e2. 22. Vilgrain V, Lewin M, Vons C, et al. Hepatic nodules in BuddeChiari syndrome: imaging features. Radiology 1999;210:443e50. 23. Cazals-Hatem D, Vilgrain V, Genin P, et al. Arterial and portal circulation and parenchymal changes in BuddeChiari syndrome: a study in 17 explanted livers. Hepatology 2003;37:510e9.
e369
24. Marin D, Galluzzo A, Plessier A, et al. Focal nodular hyperplasia-like lesions in patients with cavernous transformation of the portal vein: prevalence, MR findings and natural history. Eur Radiol 2011;21:2074e82. 25. Tanoue S, Kiyosue H, Komatsu E, et al. Symptomatic intrahepatic portosystemic venous shunt: angiographic findings and transcatheter embolisation with an alternative approach. AJR Am J Roentgenol 2003;181:71e8. 26. Bass FE, Redwine MD, Kramer LA, et al. Spectrum of congenital anomalies of the inferior vena cava: cross-sectional imaging findings. RadioGraphics 2000;20:639e52. 27. Gallego C, Miralles M, Marin C, et al. Congenital hepatic shunts. RadioGraphics 2004;24:755e72. 28. Marks C. Developmental basis of the portal venous system. Am J Surg 1969;117:671e81.
Contents lists available at SciVerse ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
Congenital portosystemic shunts associated with liver tumours L.F. Pupulim a, *, M.-P. Vullierme a, V. Paradis b, D. Valla c, S. Terraz d, V. Vilgrain a ˇ
ˇ
Department of Radiology, Hopital Beaujon, Assistance Publique des Hopitaux de Paris, Universit e Paris Diderot, Sorbonne Paris Cit e, INSERM unit e 773 CRB3, Paris, France b Department of Pathology, Hopital Beaujon, Assistance Publique des Hopitaux de Paris, Universit e Paris Diderot, Sorbonne Paris Cit e, INSERM unite 773 CRB3, Paris, France c Department of Hepatology, Hopital Beaujon, Assistance Publique des Hopitaux de Paris, Universit e Paris Diderot, Sorbonne Paris Cit e, INSERM unit e 773 CRB3, Paris, France d Department of Radiology, University Hospitals of Geneva, Geneva, Switzerland ˇ
a
ˇ
ˇ
ˇ
article in formation
AIM: To evaluate the diagnosis and presentation of liver tumours in patients with congenital portosystemic shunts (CPS). MATERIALS AND METHODS: Eight patients were diagnosed in Hopital Beaujon as having CPS. All patients underwent Doppler ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), and histological examination of liver tumours. CPS were classified according to anatomy and the amount of portal flow deviated to the systemic circulation as: total, subtotal, or partial. Liver tumours were diagnosed by needle core biopsy (n ¼ 5) or surgery (n ¼ 3). Clinical follow-up was available in all patients but one (mean follow-up 36 months; range 1e5 years). RESULTS: Six patients had total CPS, one patient had a subtotal CPS, and the last had a partial CPS. All patients presented with multiple liver nodules (range four to >15). The tumours were characterized as focal nodular hyperplasia (FNH; n ¼ 4), FNH with hepatocellular adenoma (n ¼ 2), and regenerative nodular hyperplasia (n ¼ 2). In four of seven patients (57%) that had follow-up, tumours showed enlargement or new lesions appeared. CONCLUSION: In this series of CPS patients, tumours were all benign, multiple, and of hepatocellular origin, and different tumours were present simultaneously in two patients. Tumour enlargement or new nodules were common during follow-up. Ó 2013 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. ˇ
Article history: Received 29 August 2012 Received in revised form 16 January 2013 Accepted 22 January 2013
Introduction Portosystemic shunts in patients without a history of trauma, liver biopsy, portal thrombosis, cirrhosis, or portal
* Guarantor and correspondent: L.F. Pupulim, Department of Radiology, Paris Hopital Beaujon, Assistance Publique des Hopitaux de Paris, Universite , INSERM unite 773 CRB3, 100 Blvd du Ge ne ral Diderot, Sorbonne Paris Cite Leclerc, 92110 Clichy, France. Tel.: þ33 (0) 1 40 87 53 55. E-mail address: [email protected] (L.F. Pupulim).
hypertension are considered to be congenital in origin. The latter shunts result from abnormalities in the complex embryonic development of the hepatic venous system. Recently, the extensive use of advanced imaging methods has increased the diagnosis of congenital portosystemic shunts (CPSs) and a wide variety of connections has been reported in medical literature, often as case reports. CPSs are frequently reported in association with other congenital malformations, which strengthen the embryological aetiology of such shunts. Despite being present early in the life, CPSs may remain asymptomatic until adolescence or
ˇ
ˇ
0009-9260/$ e see front matter Ó 2013 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.crad.2013.01.024
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
e363
Figure 1 A 38-year-old woman with total extra-hepatic CPS associated with multiple liver nodules diagnosed as RNH at biopsy (patient 5). (a) Axial cross-sectional T2-weighted MRI image showing the total extra-hepatic CPS. Splenic and superior mesenteric veins form a trunk that turns in the direction of the left renal vein (*). (b) CT imaging during the portal venous phase at the same level as (a) shows four ill-defined, hypo-attenuating liver nodules (black arrows). In this case, liver nodules were more conspicuous at CT than MRI.
adulthood.1,2 The morphological presentation of these shunts can show enormous variability: they can be intra- or extra-hepatic, simple or multiple, large or small, and even aneurysmal. Current classifications are based on the location of the CPS (intra- or extra-hepatic shunts) and are not related to degree of hepatic portal blood deprivation.3,4 CPSs are known to be associated with the development of
liver tumours, such as focal nodular hyperplasia (FNH),5e9 regenerative nodular hyperplasia (RNH),10,11 and hepatocellular adenoma.12,13 These tumours are considered to result from an abnormal hepatocellular response to impairment of the portal venous supply and compensatory increases in hepatic arterial flow.14e17 In the present study, the medical reports and imaging findings of eight
Figure 2 Subtotal intra-hepatic CPS in a 25-year-old woman presenting with numerous small FNH nodules and a large nodule in segment 6 diagnosed as FNH with foci of adenoma after orthotopic liver transplantation (patient 4). (a) Portography after injection of iodine contrast medium in the superior mesenteric artery shows a portosystemic shunt between the main portal vein and the atrium (black arrow). An atrophic portal branch is also seen (white arrow) defining this shunt as subtotal. (b) T1-weighted MRI image showing a hyperintense large tumour in segment 6 (white arrow). (c) Arterial-phase CT image shows that this tumour has increased in size at the follow-up 2 years later (white arrow). This was diagnosed as an FNH with internal foci of adenoma on the histopathological analysis after liver transplantation. Note also the enlarged hepatic artery (black arrow).
e364
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
Figure 3 Partial intra-hepatic CPS in an asymptomatic 23-year-old patient (patient 7). (a) Maximum intensity projection reconstructed CT image of the portal phase showing a large portosystemic shunt (*) connecting the portal trunk to the IVC. Other portal branches are present. LPV, left portal vein; RAPV, right anterior portal vein. This image also shows small calcifications that were present in an iso-attenuating FNH (black arrows) diagnosed on biopsy. (b) Cross-sectional, contrast-enhanced, axial CT image at the portal-phase at the level of the LPV. A small hypovascular non-biopsied nodule (small black arrow) and a large hypervascular nodule are seen (large black arrow). The latter was confirmed as a FNH nodule on histopathological analysis of a liver biopsy. The CPS is also seen in transverse section (*).
Table 1 Clinical setting, embryological anomalies, histopathological findings, and imaging features of liver tumours of eight patients presenting congenital portosystemic shunt (CPS). Patient
Sex; age at diagnosis
Clinical context at presentation
Laboratory findingsa
Type of portosystemic shunt
Vascular anomalies
Associated congenital anomalies
Follow-up
1
F; 24
Abdominal pain
GGT (5N); AP (2N); AST/ALT (1,5 N)
Total, extrahepatic, single
Trunk SMV þ SV to IVC
No
2
F; 14
No symptoms
GGT (4N); AP (5N); AST/ALT (3N)
Total, extrahepatic, single
Trunk SMV þ SV to right atrium
3
M; 13
Right upper quadrant pain
GGT (3N); AP (2N); bilirubin (1,2N); AST/ALT (1,5 N).
Total, intrahepatic, single
RPV to RHV through an aneurysm. No other portal branches
Ataxiatelangiectasia syndrome; intestinal malrotation Turner syndrome; aorta coarctation No
4
F; 25
No symptoms
GGT (2N)
Subtotal, intrahepatic, single
LPV to right atrium through a PDV. No right portal vein
No
5
F; 38
Abdominal pain
Normal values
F; 16
Right upper quadrant pain
7
F; 23
Abdominal pain
GGT (4N); AP (2N); serum ammonia (1,5N) Normal values
Trunk SMV þ SV to left renal vein Large tube connecting SV to left renal vein
No
6
Total, extrahepatic, single Total, extrahepatic, single Partial, intrahepatic, single
RPV to IVC in the confluence of RHV
8
F; 24
No symptoms
Normal values
Total, intrahepatic, single
RPV to IVC without other portal branches
Lobar pulmonary sequestration No
No
60 months e Tumour enlargement from 7 to 13 cm 36 months e Surgery for tumours that presented enlargement 27 months e Surgery for tumours that presented enlargement 41 months e no changes 59 months e no changes 9 months - multiples small enhancing nodules appeared 24 months e no changes
F, female; M, male; AP, serum alkaline phosphatase; GGT, serum gamma-glutamyl-transferase activity; AST/ALT: transaminases; SMV, superior mesenteric vein; SV, splenic vein; IVC, inferior vena cava; RPV, right portal vein; RHV, right hepatic vein; LPV, left portal vein; PDV, persistent ductus venosus; OLT, orthotopic liver transplantation, US, ultrasound; NE, non-enhanced; CE, contrast enhanced; CT, computed tomography; MRI, magnetic resonance imaging; NCB, needle core biopsy; FNH, focal nodular hyperplasia; RNH, regenerative nodular hyperplasia; IV, intravenous; nod, nodule. a Only abnormal results are mentioned in laboratory findings. Numbers in parentheses express the times serum levels were above upper normal values (N). b After OLT the pathological liver analysis showed multiples FNH not identified on imaging.
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
consecutive patients with CPS were retrospectively analysed with the aim of evaluating the diagnosis and presentation of hepatic tumours in patients with CPS.
Material and methods During a 15 year period (1993e2008) eight patients were diagnosed in Hopital Beaujon with CPS associated with liver tumours. The medical data, imaging findings, and histopathology findings were retrospectively reviewed. There was no history of abdominal surgery, trauma, or liver biopsy before the diagnosis of the shunt. Cirrhosis, chronic hepatitis, portal vein thrombosis, and portal hypertension were ruled out in all patients following assessment of the imaging, clinical, and laboratory findings. Serum alpha-fetoprotein and carcinoembryonic antigen were normal in all patients. Clinical follow-up was available in all patients but one. The mean total follow-up was 36 months (range 1e5 years). ˇ
Imaging technique Doppler ultrasound, abdominal multiphasic computed tomography (CT), and magnetic resonance (MR) imaging were performed in all patients. Five patients also underwent
e365
selective hepatic angiography of liver vessels. Angiography was not performed in the other three patients because of vascular reconstructions available on CT and/or MRI that were considered satisfactory to evaluate hepatic vascular anatomy. The present study was a retrospective analysis of examinations performed throughout a period of 15 years; as a result, protocols of the CT and MRI examinations were not uniform for all eight patients. Different CT machines were used (CT Twin Flash; Marconi Medical Systems, Haifa, Israel; CT Lightspeed with 8 and 64 detectors, GE Healthcare, Milwaukee, WI, USA). Non-ionic intravenous contrast material (120e160 ml; 350 mg iodine/ml) was administered using an automatic power injector. Arterial and portal venous acquisitions were performed. MRI was performed using two different 1.5 T units (Elscint, Haifa, Israel) and (Intera, Philips Medical Systems, Best, The Netherlands) with acquisition of T1 and T2-weighted images in gradient echo (GE) and/or fast spin-echo techniques. Contrast-enhanced T1-weighted GE images were obtained after intravenous injection of 0.1 mmol/kg of body weight of non-specific chelates of gadolinium. Digital hepatic angiography (Multistar T.O.P.; Siemens; Munich, Germany) examinations were performed by catheterization of the coeliac trunk using the Seldinger technique. Arterial and venous phases were performed, the
Outcome
Number of nodules on imaging
Nodules size range (diameter in cm)
Pathologic proof
Tumour diagnosis
NE-CT (attenuation)
MRI T1 (intensity)
MRI T2 (intensity)
Enhancement on arterial phase of IV contrast
No treatment performed
6 (3 large and 3 small)
1e10
NCB (2 largest nodules)
FNH
Hypo, heterogeneous
Iso (large nodules: centre hypo)
Heterogeneous; hyper in centre (large nodules)
Yes
Clinically stable
>15
1e13
Surgical biopsy (3 nodules)
FNH
Iso (largest nodule: hypo)
Hyper
Hypo
Yes
2e15
Surgery (OLT)
Hypo
Hyper (2 nod); Hypo (2 nod)
No
1e12
Surgery (OLT)
Hypo
Hyper (2 nod); Hypo (2 nod) Hyper
Iso; slightly hyper
Yes (heterogeneous)
Hypo, iso
Iso
Hyper, iso
No
4b
Clinically stable
>15
1e5
NCB (1 nod)
One nodule: adenoma Other nodules: FNH. FNH with foci of adenoma (large nodules). Small nodules: FNH. RNH
Clinically stable
>15
1e4
NCB (2 nod)
RNH
Hypo, iso
Hyper
Hyper
Yes
Clinically stable
>10
1e7
NCB (2 nod)
FNH
Hypo
Iso
Iso
Yes
Clinically stable
6
1e4
NCB (1 nod)
FNH
Hypo
Hyper
Iso; Hyper (largest)
Yes (2 largest); No (others)
OLT for severe encephalopathy after tumour resections OLT for severe encephalopathy after tumour resections
>8
e366
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
latter indirectly evaluated the portal vein and the portosystemic shunts after injection of iodine-containing contrast medium in the superior mesenteric and/or splenic artery.
Imaging analysis Images were retrospectively reviewed by two experts in hepatic imaging in consensus and CPSs were defined according to the degree of hepatic portal blood deprivation as follows: (a) total: when all portal blood flow was seen to be deviated to the systemic circulation on imaging and no intrahepatic portal branches could be identified (Fig 1); (b) subtotal: most of the portal blood flow was deviated to the systemic circulation but some abnormal atrophic intra-hepatic portal branches could be identified on imaging (Fig 2); (c) partial: besides the shunt, normal intra-hepatic portal branches could be identified, meaning that some of the portal blood still passed through the liver sinusoids (Fig 3). Moreover, according to the degree of portal blood deprivation, total and subtotal CPSs were considered as major. CPSs were also defined as intra-hepatic when the abnormal connection was located within the liver and as extra-hepatic when the shunt had no relation to the liver. The number of shunt pathways was also assessed to define them as single or multiple. Liver tumours were analysed according to: (a) number of tumours, (b) size, (c) attenuation on unenhanced CT, (d) lesion intensity on T1- and T2-weighted MRI sequences, and (e) enhancement on contrast-enhanced studies (Table 1). Typical imaging findings for FNH,15 hepatocellular adenoma,18,19 and RNH17 were analysed. For FNH, imaging findings included: lesion homogeneity; intense and homogeneous contrast enhancement on arterial-phase imaging with lack of wash-out; central scar; central arteries on Doppler ultrasound; and absence of capsule. For telangiectatic/inflammatory hepatocellular adenomas, imaging findings included: marked hypersignal on T2-weighted sequences; strong arterial enhancement with persistent enhancement in portal and delayed phases; and absence of signal dropout on chemical shift sequences. For steatotic hepatocellular adenomas, imaging findings included: signal dropout on T1-weighted chemical shift sequence; isosignal or slight hypersignal on T2-weighted images; and moderate enhancement in the arterial phase with no persistent enhancement in the portal and delayed phases. The diagnosis of RNH was suspected when a lesion was of high signal intensity on T1-weighted images and iso- or slightly hypersignal intensity on T2-weighted images. Tumour enlargement and the appearance of new tumours were determined during follow-up. The hepatic artery was also evaluated, and it was considered dilated when the diameter at the origin was greater than that of the splenic artery.
Histopathological analysis Needle core biopsy (NCB) was performed in all patients sampling either one or two nodules (Table 1). As patients presented with multinodular liver disease, target nodules were chosen according to the possibilities of best access. Biopsy of the adjacent liver was also obtained in all patients.
One patient had a corresponding surgical biopsy where three nodules were sampled. In two patients, additional tumour cell clonal tests were performed and further histopathological analysis was obtained after orthotopic liver transplantation (OLT). Histopathological records of liver tumours and liver parenchyma were retrospectively reviewed by an experienced liver pathologist. Tumour nodules were reviewed after haematoxylin and eosin (H&E), trichrome, and reticulin staining. FNH was considered when the lesion was well-circumscribed and composed of benign-appearing hepatocytes arranged in nodules that were partly or completely delineated by fibrous septa. Dystrophic vessels were observed inside the fibrous septa, focally associated with ductular proliferation. The hepatocytes were hyperplastic, arranged in liver plates of normal or slightly increased thickness with a well-preserved reticulin framework. When the hepatocellular component appeared more proliferative and extensive, a diagnosis of FNH with adenomatous features was considered. RNH was defined as nodular architecture of the liver parenchyma without extensive fibrosis, showing areas of thickened liver cell plates associated with areas of atrophic cell plates. Liver samples without tumour were histologically assessed after H&E and trichrome staining. The extent of portal fibrosis, presence of sinusoidal dilatation, necroticoinflammatory changes, and intensity of steatosis were systematically evaluated.
Results Clinical findings The patients included seven women and one man with a mean age of 22-years-old (range 13e38 years). Three patients had incidental discovery of CPSs and five presented with abdominal pain. An associated congenital anomaly was found in three patients as follows: ataxia-telangiectasia syndrome and intestinal malrotation (n ¼ 1); Turner’s syndrome and coarctation of the aorta (n ¼ 1), and lobar pulmonary sequestration (n ¼ 1). A summary of the clinical findings, liver blood tests, histopathological findings, and imaging features are shown in Table 1.
CPSs A CPS was seen during Doppler ultrasound in all patients. Contrast-enhanced CT confirmed the presence of portosystemic shunts in all patients. Six patients had a total CPS; four were extra-hepatic and these shunts were connected to the inferior vena cava (IVC; n ¼ 1), the left renal vein (n ¼ 2), or the right atrium (n ¼ 1). Among the two total intra-hepatic portosystemic shunts, one presented as an aneurysm connection between the right portal vein and the right hepatic vein, and the other connected the right portal vein to the IVC. One patient had a subtotal CPS with an atrophic intra-hepatic portal branch (Fig 2). One patient had a partial CPS between the right branch of the portal vein and the IVC (Fig 3). The intra-hepatic portal branches were normal in this patient. All shunts had a single pathway
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
defining an individual shunt. The hepatic artery was dilated in all patients with total and subtotal shunts (n ¼ 7) with a diameter ranging from 7e9 mm. It was considered of normal calibre in the patient with a partial shunt.
Liver nodules All eight patients presented with multiple liver tumours ranging from four to greater than 15 nodules. Four patients had FNH, two had RNH, one (patient 3) presented multiple FNH and a single hepatocellular adenoma, and another patient (patient 4) was diagnosed with multiple FNH nodules and three of the largest FNH nodules were associated with foci of hepatocellular adenoma. Imaging features of liver tumours are summarized in Table 1. Among the four patients with FNH, only two (patients 1 and 8) had all imaging features of typical FNH15 (Fig 4). The liver adenoma and the FNH nodules with foci of liver adenoma were hypo-attenuating on non-enhanced CT images and presented with heterogeneous enhancement on arterial-phase imaging. The nodules in these patients (patients 3 and 4) did not show typical findings of FNH either of telangiectatic/inflammatory or steatotic hepatocellular adenoma. These two patients underwent tumour resection because of a preoperative diagnosis of hepatocellular adenoma on percutaneous biopsy and tumours enlarged during
e367
follow-up. After surgery they experienced a degree of hepatic encephalopathy and persistent hepatic failure and finally underwent liver transplantation. Typical imaging features of RNH were found in one of two patients. One patient was lost to follow-up. Three patients presented with nodule enlargement during followup. One patient (patient 7), with four nodules at the time of diagnosis, developed multiple small hypervascular liver nodules during the follow-up. The tumours remained stable in the other three patients.
Discussion The current series of patients with CPS analyses both the variety of liver tumours associated with CPS as well as the morphological aspects of these interesting congenital shunts. Liver tumours, such as FNH, are known to develop in response to focal or diffuse abnormal perfusion of liver.5e17,20e24 In the normal liver, they are usually solitary, whereas they are usually multiple in patients with vascular disorders such as BuddeChiari syndrome, congenital hepatic fibrosis, OslereRendueWeber disease, and cavernous transformation of the portal vein. Similarly, FNH have been described in association with CPSs.5e9 FNH were by far the most common tumours observed in this series. CPS results in
Figure 4 Total intrahepatic CPS in a 24-year-old woman (patient 8). (a) Arterial-phase T1-weighted MRI image showing two enhanced liver nodules (black arrows). CPS is seen in transverse section (*). (b) Portal-phase, T1-weighted MRI image (more cranial section) shows a typical enhanced FNH of the liver dome (black arrow) with a lobulate contour and central scar. Diagnosis was confirmed by biopsy. CPS (*) is almost joining the IVC at this level. (c) T2-weighted MRI image of the same nodule as (b) showing a hyperintense central scar within the FNH (black arrow).
e368
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369
portal venous deprivation, which is often complete, and this induces a compensatory increase in arterial inflow. Indeed, arterial compensation was confirmed in all cases with major portal deprivation (either total or subtotal CPS). The only patient who did not have an enlarged hepatic artery had a partial shunt, suggesting that haemodynamic changes are more prominent in major shunts than in partial ones. The present series also confirms the possible association of other hepatocellular lesions with CPS, such as hepatocellular adenoma and RNH. Two of the present patients (25%) presented simultaneously with FNH and hepatocellular adenoma. The diagnosis in the case of FNH with foci of hepatocellular adenoma was confirmed by immunohistochemical analysis and tumour cells clonal pattern. The mechanisms that lead to the development of monoclonal tumours within polyclonal nodules have to be elucidated. Hence, CPS may play a role in such cases. In view of the possibility of malignant transformation of hepatocellular adenoma, patients with CPS should be closely followed up. In the present series, all patients were adults and had multiple liver tumours. Hepatologists and radiologists should be alerted to the possibility of CPS in young adults presenting with spontaneous multiple liver tumours. Imaging, and especially MRI, plays an important role in diagnosing FNH. The reported sensitivities and specificities are 70% and 98%, respectively.15 In the present series of patients with CPS, FNHs were atypical in 50% of patients (two out of four). It is possible that complete and chronic portal deprivation could alter the morphology of FNH. The variety of lesions that are encountered in patients with CPS should reinforce the role of liver biopsy, notably in patients with atypical lesions at imaging. In four of seven patients with serial imaging, liver tumours enlarged in size or new nodules appeared. To the authors’ knowledge, this is the first such description and suggests that CPS may be a progressive disease. Follow-up and treatment, as well as CPS closure, may be considered for these patients.25 Excess involution and abnormal connections of the periintestinal vitelline venous network are involved in the origin of CPS.26e28 The diversity of CPS in terms of size, localization, number of pathways, and possible association with normal portal intra-hepatic anatomy is related to the complex development of the embryonic hepatic venous system. Previous classifications based on morphology were proposed and separate shunts according to their level, either intra-hepatic3 or extra-hepatic.4 However, these methods fail to classify all cases and do not evaluate the degree of deprivation of portal blood flow. The morphology of CPS could be described according to the presence and degree of intra-hepatic portal venous flow as highlighted here. This classification may better explain the relationship between CPS and hepatocellular lesions. Most of the present patients with hepatocellular lesions had a major shunt (either total or subtotal). One limitation of the present study was the small number of patients, although all clinical and morphological details were available. Although CPS is rare, further studies with larger number of patients are necessary to evaluate progressive changes in liver tumours in patients with CPS.
In conclusion, although CPS is a congenital disease, in the present study a number of patients were diagnosed in adulthood with a number of benign hepatocellular tumours. These lesions were all benign and multiple. Varying types of tumour (FNH and hepatocellular adenoma) may coexist in the same patient. During follow-up, tumour enlargement or new nodules appeared in over half of the present patients.
Acknowledgements The authors thank the substantial assistance of Dr Dermot O’Toole (Department of Clinical Medicine & Gastroenterology of St James’s Hospital & Trinity College Dublin, Ireland) for editing the manuscript.
References 1. Kakitsubata Y, Kakitsubata S, Kiyomizu H, et al. Intrahepatic portalhepatic venous shunts demonstrated by US, CT and MR imaging. Acta Radiol 1996;37:680e4. 2. Murray CP, Yoo SJ, Babyn P. Congenital extrahepatic portosystemic shunts. Pediatr Radiol 2003;33:614e20. 3. Park JH, Cha SH, Han JK, et al. Intrahepatic portosystemic venous shunt. AJR Am J Roentgenol 1990;155:527e8. 4. Morgan G, Superina R. Congenital absence of the portal vein: two cases and a proposed classification system for portosystemic vascular anomalies. J Pediatr Surg 1994;29:1239e41. 5. Altavilla G, Cusatelli P. Ultrastructural analysis of the liver with portal vein agenesis: a case report. Ultrastruct Pathol 1998;22:477e83. 6. Guariso G, Fiorio S, Altavilla G, et al. Congenital absence of the portal vein associated with focal nodular hyperplasia of the liver and cystic dysplasia of the kidney. Eur J Pediatr 1998;157:287e90. 7. Tanaka Y, Takayanagi M, Shiratori Y, et al. Congenital absence of portal vein with multiple hyperplastic nodular lesions in the liver. J Gastroenterol 2003;38:288e94. 8. Kim T, Murakami T, Sugihara E, et al. Hepatic nodular lesions associated with abnormal development of the portal vein. AJR Am J Roentgenol 2004;183:1333e8. 9. De Gaetano AM, Gui B, Macis G, et al. Congenital absence of the portal vein associated with focal nodular hyperplasia in the liver in an adult woman: imaging and review of the literature. Abdom Imaging 2004;29:455e9. 10. Grazioli L, Alberti D, Olivetti L, et al. Congenital absence of portal vein with nodular regenerative hyperplasia of the liver. Eur Radiol 2000;10:820e5. 11. Arana E, Martí-Bonmatí L, Martínez V, et al. Portal vein absence and nodular regenerative hyperplasia of the liver with giant inferior mesenteric vein. Abdom Imaging 1997;22:506e8. 12. Nakasaki H, Tanaka Y, Ohta M, et al. Congenital absence of the portal vein. Ann Surg 1989;210:190e3. 13. Kamiya S, Taniguchi I, Yamamoto T, et al. Analysis of intestinal flora of a patient with congenital absence of the portal vein. FEMS Immunol Med Microbiol 1993;7:73e80. 14. Wanless IR, Lentz JS, Roberts EA. Partial nodular transformation of liver in an adult with persistent ductus venosus. Arch Pathol Lab Med 1985;109:427e32. 15. Hussain SM, Terkivatan T, Zondervan PE, et al. Focal nodular hyperplasia: findings at state-of-the-art MR imaging, US, CT, and pathologic analysis. RadioGraphics 2004;24:3e19. 16. Wanless IR, Mawdesley C, Adams R. On the pathogenesis of focal nodular hyperplasia of liver. Hepatology 1985;5:1194e200. n M, Pe rez A, et al. Clinical and radiological 17. Alonso-Gamarra E, Parro manifestations of congenital extrahepatic portosystemic shunts: a comprehensive review. RadioGraphics 2011;31:707e22. 18. Laumonier H, Bioulac-Sage P, Laurent C, et al. Hepatocellular adenomas: magnetic resonance imaging features as a function of molecular pathological classification. Hepatology 2008;48:808e18.
L.F. Pupulim et al. / Clinical Radiology 68 (2013) e362ee369 19. Ronot M, Bahrami S, Calderaro J, et al. Hepatocellular adenomas: accuracy of magnetic resonance imaging and liver biopsy in subtype classification. Hepatology 2011;53:1182e91. 20. Wanless IR. Micronodular transformation (nodular regenerative hyperplasia) of the liver: a report of 64 cases among 2500 autopsies and a new classification of benign hepatic nodules. Hepatology 1990;11:787e97. 21. Weinbren K, Washington SLA. Hyperplastic nodules after portacaval anastomosis in rats. Nature 1976;264:440e2. 22. Vilgrain V, Lewin M, Vons C, et al. Hepatic nodules in BuddeChiari syndrome: imaging features. Radiology 1999;210:443e50. 23. Cazals-Hatem D, Vilgrain V, Genin P, et al. Arterial and portal circulation and parenchymal changes in BuddeChiari syndrome: a study in 17 explanted livers. Hepatology 2003;37:510e9.
e369
24. Marin D, Galluzzo A, Plessier A, et al. Focal nodular hyperplasia-like lesions in patients with cavernous transformation of the portal vein: prevalence, MR findings and natural history. Eur Radiol 2011;21:2074e82. 25. Tanoue S, Kiyosue H, Komatsu E, et al. Symptomatic intrahepatic portosystemic venous shunt: angiographic findings and transcatheter embolisation with an alternative approach. AJR Am J Roentgenol 2003;181:71e8. 26. Bass FE, Redwine MD, Kramer LA, et al. Spectrum of congenital anomalies of the inferior vena cava: cross-sectional imaging findings. RadioGraphics 2000;20:639e52. 27. Gallego C, Miralles M, Marin C, et al. Congenital hepatic shunts. RadioGraphics 2004;24:755e72. 28. Marks C. Developmental basis of the portal venous system. Am J Surg 1969;117:671e81.