Intrahepatic cholangiocarcinoma (ICC) is a highly

Transcription

1 Intrahepatic Peripheral Cholangiocarcinoma in Cirrhosis Patients May Display a Vascular Pattern Similar to Hepatocellular Carcinoma on Contrast-Enhanced Ultrasound Ramón Vilana, 1,2 * Alejandro Forner, 2,3 * Luis Bianchi, 1,2 Ángeles García-Criado, 1 Jordi Rimola, 1,2 Carlos Rodríguez de Lope, 3 María Reig, 2,3 Carmen Ayuso, 1,2 Concepció Brú, 1,2 and Jordi Bruix 2,3 The aim of this study was to describe the imaging features by contrast-enhanced ultrasound (CEUS) of intrahepatic cholangiocarcinoma (ICC) in cirrhosis patients. We registered the CEUS images of cirrhosis patients with histologically confirmed ICC. In all cases magnetic resonance imaging (MRI) was done to confirm the diagnosis and/or staging purposes. A total of 21 patients met all the criteria to be included in the study. The median nodule size was 32 mm. All nodules showed contrast enhancement at arterial phase; in 10 cases it was homogeneous and in 11 cases peripheral (rim-like). All nodules displayed washout during the venous phases; it appeared during the first 60 seconds in 10 nodules, between seconds in five cases, and in six cases after 2 minutes. Ten nodules (five larger than 2 cm) displayed homogeneous contrast uptake followed by washout and they correspond to the specific pattern of hepatocellular carcinoma according to the American Association for the Study of Liver Diseases criteria. However, none of these lesions displayed washout on MRI. Conclusion: CEUS should not be used as the sole imaging technique for conclusive hepatocellular carcinoma diagnosis and if the MRI does not display the diagnostic vascular pattern, a confirmatory biopsy is mandatory. (HEPATOLOGY 2010;51: ) Intrahepatic cholangiocarcinoma (ICC) is a highly malignant epithelial cancer originating from the bile ducts with features of cholangiocyte differen- Abbreviations: CEUS, contrast-enhanced ultrasound; CT, computed tomography; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; ICC, intrahepatic cholangiocarcinoma; MRI, magnetic resonance imaging. *These authors contributed equally to this work. From the 1 Radiology Department, Barcelona Clinic Liver Cancer (BCLC) group, Hospital Clinic Barcelona, IDIBAPS, University of Barcelona, Barcelona, Spain; 2 Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Barcelona, Spain; and 3 Liver Unit, Barcelona Clinic Liver Cancer (BCLC) group, Hospital Clinic Barcelona, IDIBAPS, University of Barcelona, Barcelona, Spain. Received November 28, 2009; accepted January 22, BCLC is funded through the Spanish Biomedical Research Network (CIBER) for the area of Hepatic and Digestive disorders. This work was supported in part by grants of the Instituto de Salud Carlos III (grant PI 06/ 132 and PI 08/0146). A.F. is partially supported by a grant of the Instituto de Salud Carlos III (PI 05/645). M.R. is funded by a grant of the BBVA foundation. C.R.d.L. is partially supported by a grant of the Instituto de Salud Carlos III (PI09/510) and funded by a grant of the BBVA foundation. Address reprint requests to: Ramon Vilana, BCLC group, Department of Radiology-CDI, C/ Villarroel 170, Barcelona, Spain; fax: þ rvilana@clinic.ub.es. Copyright VC 2010 by the American Association for the Study of Liver Diseases. Published online in Wiley InterScience ( DOI /hep Potential conflict of interest: Nothing to report. tiation. 1,2 ICC constitutes the second most common primary liver tumor and epidemiologic studies suggest its incidence is increasing in Western countries. 3 Several risk factors for ICC development have been suggested. The most known are primary sclerosing cholangitis, hepatobiliary flukes, hepatolithiasis, and biliary malformations. In addition, cirrhosis, mainly secondary to chronic infection with hepatitis C virus (HCV), has been recognized as an important risk factor for ICC development and this association has been speculated as the cause of the increasing incidence of ICC in the last years. 4,5 It is well known that the most frequent malignancy in cirrhosis is hepatocellular carcinoma (HCC). However, in recent years ICC has been found in patients with cirrhosis and its correct differentiation from HCC is a major clinical issue because the prognosis and treatment options differ significantly. 4,6 In cirrhosis patients, surveillance for early HCC diagnosis is widely accepted and relies on biannual abdominal ultrasound (US). 7,8 Upon detection of a suspicious nodule a recall strategy based on dynamic imaging and/or biopsy should be initiated. 8 Major information is available to characterize the HCC 2020

2 HEPATOLOGY, Vol. 51, No. 6, 2010 VILANA ET AL dynamic imaging pattern after contrast-enhanced US (CEUS), computed tomography (CT), and magnetic resonance imaging (MRI): intense arterial uptake followed by washout in the venous phase by dynamic imaging techniques. This has allowed the definition of noninvasive imaging criteria to diagnose HCC in the setting of liver cirrhosis. 8,9 However, very few studies have described the imaging appearance of ICC by CEUS, and in most cases they included large tumors in noncirrhosis patients For that reason there is no accurate information regarding the appearance of ICC after CEUS in cirrhosis patients, and particularly in small nodules detected during screening. In this setting the diagnostic difficulties may be higher and, as previously mentioned, it is key to distinguish between HCC and ICC to permit a proper treatment indication. The purpose of this study was to assess the CEUS features of ICC in the setting of cirrhosis, taking into account the nodule size and making special emphasis on the enhancement pattern and the differential diagnosis with HCC. Secondarily, we compared the findings on CEUS with those obtained by MRI, highlighting the differences in the contrast enhancement in each phase. Patients and Methods Patients. We registered the radiological images of all patients with ICC and cirrhosis consecutively diagnosed in our institution between December 2003 and September The inclusion criteria were as follows: (1) patient with cirrhosis diagnosed by imaging, pathology, and/or clinical criteria (presence of portal hypertension measured by hepatic vein catheterization or presence of esophageal varices at endoscopy); and (2) pathology-confirmed diagnosis of ICC by percutaneous biopsy or surgical specimen analysis. Patients with mixed hepatocellular-cholangiocarcinoma were excluded. In those cases with multiple nodules, only the larger lesion with histological confirmation was analyzed. The CEUS and MRI studies were performed for diagnosis and staging purposes within less than a month. Image Acquisition: CEUS. US examinations were performed using gray-scale US Sequoia 512 equipment (Siemens, Mountain View, CA). Baseline study in the fundamental mode was performed with a multifrequency 4C1 convex array probe. The gray-scale US characteristics of the tumor, including size, location (lobule, segment, depth), and echogenicity, were recorded. CEUS was performed using coherent-contrast imaging (CCI) with the same convex array probe. The focus depth was placed beyond the area of interest and using the following settings: insonating frequency, 3 MHz; acoustic power, 75 to 90 db; frame rate, A low mechanic index (<0.2) was selected to avoid disruption of microbubbles. CEUS studies were performed after the administration of 2.4 ml bolus of SonoVue (Bracco, Italy) through a 21G peripheral intravenous cannula, followed by a 5-mL saline flush. This bolus was repeated if the first evaluation was not technically adequate. Lesion enhancement patterns were studied for 3.5 minutes to register arterial (10-30 seconds), portal ( seconds), and late phases (>120 seconds). Image Acquisition: MRI. MRI was performed in all patients with a 1.5-T MRI system (Symphony, Siemens Medical Systems, Erlangen, Germany; and SIGNA CVi, General Electric Medical System, Milwaukee, WI) using a phased-array coil for signal detection. All patients underwent transverse T1-weighted in-phase (100/5.24 repetition time / echo time [TR/ TE], matrix, 70 flip angle) and out-ofphase gradient echo (100/2.38 TR/TE, matrix, 70 flip angle), slice thickness of 5 mm, transverse T2-weighted breath-hold, slice thickness of 5 mm (1100/116 TR/TE, matrix), and axial dynamic multiphasic contrast-enhanced 3D of the liver with fat suppression before and after intravenous contrast administration (volumetric interpolated breathhold examination [VIBE] sequence in cases of Symphony; and 3D breath-hold fast spoiled gradient echo in cases of SIGNA) with a slice thickness of mm. Gadolinium (gadodiamide 0.5 mmol/l Ominscan-Amersham, Madrid, Spain) was injected at a dose of 0.2 ml/kg at a rate of 2 ml/s, followed by a flush of 20 ml saline. A bolus tracking technique was used to obtain arterial-phase images, 20 seconds after contrast injection. Portal venous and delayed venous phase images were acquired and seconds thereafter. Finally, a T1-weighted 2D gradient echo with fat suppression MRI was performed (160/2.6 TR/TE, matrix) 5 minutes after contrast injection. Image Analysis. Images on CEUS were videotaped and analyzed by at least two radiologists with more than 8 years of experience in abdominal radiology (R.V., L.B., A.G.C., and C.B.). Images on MRI were independently analyzed by two abdominal radiologists with 5 and more than 10 years of experience (J.R. and C.A., respectively). The nodule appearance on US was registered as hyperechoic, isoechoic, or hypoechoic

3 2022 VILANA ET AL. HEPATOLOGY, June 2010 compared to surrounding parenchyma at baseline and as hypointense, isointense, hyperintense, or heterogeneous on precontrast T1-weighted and T2-weighted MRI sequences. The contrast behavior in each phase was analyzed compared to the surrounding parenchyma and was categorized as: (1) homogeneously hyperenhanced, when the whole nodule shows contrast uptake; (2) partially hyperenhanced, when the nodule enhancement involves more than 25% of the lesion, except the central area; (3) peripherally hyperenhanced, when the enhancement was limited to less than 25% of the periphery of the lesion; (4) heterogeneously hyperenhanced, when the nodule displayed mixed irregular areas of contrast enhancement; (5) isoenhanced: the lesion enhances in the similar degree as the surrounding parenchyma; and (6) hypoenhanced: the lesion enhances in the less degree than the surrounding parenchyma. The presence of contrast washout and its time of occurrence were registered (during the first 60 seconds, between seconds, and after 120 seconds following contrast administration). Statistical Analysis. Data were analyzed on a bylesion basis. Quantitative variables were expressed as median and range and categorical as count and proportions. Differences in the basal and specific contrast enhancement in each phase according to nodule size (2 cm, 2-3 cm, and >3 cm) were evaluated by chisquare test / Fisher s exact test. A P-value less than 0.05 was considered significant. Calculations were done with SPSS v. 16 (SPSS, Chicago, IL). Results Patient Characteristics. A total of 21 patients (12 men and 9 women, median age 63, range years) met all the criteria to be included in the study. Fifteen patients (71.4%) had a uninodular ICC, four patients (19%) had two nodules, one patient (4.8%) three nodules, and one patient (4.8%) six nodules. In the six cases with more than one nodule, all of them were located in a different segment, and were thus not considered as a peripheral satellite nodule. The etiologies of liver cirrhosis were chronic HCV infection (13 patients), chronic HCV infection combined with alcohol abuse (two patient), alcohol abuse (two patients), hemochromatosis (two patient), and cryptogenetic (two patients). All patients except one had a Child- Pugh score A. The diagnostic work-up was initiated because of abdominal pain in one patient, constitutional syndrome in two cases, and in the setting of screening US for early HCC diagnosis in 18 cases (85.7%). The pathological diagnosis of ICC was done Table 1. Main Patient and Tumor Characteristics Age Median (Years) 63 [49-78] Male/Female (n) 12/9 Etiology (n): HCV 13 Ethanol 2 Hemochromatosis 2 HCV þ ethanol 2 Cryptogenetic 2 Child-Pugh A/B (n) 20/1 ASAT median (UI/L) 54 [21-149] ALAT median (UI/L) 55 [15-167] AP median (UI/L) 186 [ ] GGT median (UI/L) 63 [15-290] PR median (%) 89 [61-100] Bilirubin median (mg/dl) 1 [ ] Platelets median (10 9 /L) 151 [51-262] Albumin median (g/dl) 42 [24-47] AFP median (ng/ml) 5.5 [1-54] Nodule size median (mm) 32 [15-70] 20 mm (n) mm (n) 6 >30 mm (n) 11 Number of nodules: 1/2/3/>3 (n) 15/4/1/1 AFP, alpha-fetoprotein; ASAT, aspartate aminotransferase; ALAT, alanine aminotransferase; GGT, gamma-glutamyl transpeptidase; AP, alkaline phosphatase; HCV, hepatitis C virus; HCB, hepatitis B virus; PBC, primary biliary cirrhosis; PR. prothrombin ratio. by ultrasonography-guided percutaneous liver biopsy in 20 patients and by pathology examination of the resected specimen in one patient with suspected ICC by imaging and initial negative percutaneous biopsy. In all six cases treated by surgical resection after a positive biopsy the ICC was confirmed by the analysis of the resected specimen. The main patient characteristics are summarized in Table 1. CEUS Findings. The median nodule size was 32; range mm. Four out of 21 nodules were 2 cm (19%), six were 2-3 cm (28.6%), and 11 (52.4%) were larger than 3 cm. Eighteen out of the 21 nodules (85.7%) were hypoechoic and three (14,3%) heterogeneous on the baseline study. In the arterial phase all nodules displayed contrast uptake. In 10 cases (47.6%) the nodule was homogeneously hyperenhanced (Figs. 1, 2) and in the other 11 cases (52.4%), the enhancement was peripheral (Fig. 3). During the portal phase, six nodules (28.6%) were isoenhanced and 15 (71.4%) hypoenhanced compared to the surrounding parenchyma. Finally, in the delayed phases all nodules were hypoenhanced, and therefore displayed contrast washout (Figs. 1-3). This contrast washout was evidenced during the first 60 seconds in 10 cases (47.6%), between 60 and 120 seconds in 5 nodules (23.8%), and after 120 seconds in six cases (28.6%). There were no statistically significant differences in baseline pattern and enhancement in arterial, portal, and delayed

4 HEPATOLOGY, Vol. 51, No. 6, 2010 VILANA ET AL Fig. 1. ICC in a 64-year-old man with cirrhosis secondary to HCV. (A) Baseline ultrasonography: 25 mm solid nodule at V hepatic segment. (B) Arterial phase: complete homogeneous enhancement of the nodule (30 seconds). (C,D) Portal and late phases: washout at 60 and 120 seconds. phases according to the nodule size. However, the nodules smaller than 2 cm were isoenhanced in portal phase in 75% of cases, compared to only in 17.6% of nodules larger than 2 cm. In addition, the washout appeared later in smaller nodules; in that sense, in three out of four nodules less than 2 cm (75%) the washout appeared beyond 120 seconds compared to only three nodules (17.6%) larger than 2 cm. Table 2 and Fig. 4 summarize the main results on CEUS. MRI Findings. On T1-weighted sequences, 18 lesions (85.7%) appeared as hypointense, two nodules (9.5%) as isointense, and the remaining nodule as hyperintense (4.8%). On T2-weighted images, 13 nodules (61.9%) appeared as hyperintense, four (19%) as isointense relative to the hepatic parenchyma, two (9.5%) as heterogeneous mass, and finally two (9.5%) appeared as hypointense. On arterial phase, all lesions displayed contrast uptake: in 19 cases (90.5%) peripheral (Fig. 3) and in two cases the contrast enhancement was homogeneous (9.5%) (Fig. 2). On portal venous phase, nine nodules (42.9%) displayed a partial lesion enhancement, 10 nodules (47.6%) a peripheral enhancement, and in two lesions (9.5%) the enhancement was homogeneous. Finally, on delayed phase, 11 nodules (52.4%) had a homogeneous enhancement (Fig. 2), eight lesions (38.1%) a partial enhancement, and two nodules (9.5%) displayed a peripheral contrast enhancement (Fig. 3). None of the ICC nodules were iso- or hypoenhanced compared with the surrounding liver parenchyma during all phases of dynamic study. There were no statistically significant differences in baseline pattern and enhancement in arterial, portal, and delayed phases according to nodule size, although the smaller nodules tend to appear homogeneously hyperenhanced on delayed phase more frequently than larger lesions (Table 3).

5 2024 VILANA ET AL. HEPATOLOGY, June 2010 Fig. 2. ICC in a 75-year-old man with cirrhosis secondary to HCV. (A) Baseline ultrasonography: 15 mm hypoechoic nodule at VI hepatic segment. (B) Arterial phase: complete homogeneous enhancement. (C) Late phase: Washout at 120 seconds. (D,E) MRI of the same patient that displays homogeneous enhancement of intravenous contrast on arterial phase (D) and delayed phase (E) of the dynamic sequences. Degree of Concordance Between CEUS and MRI. On arterial phase all nodules were hyperenhanced compared with the surrounding liver parenchyma in both imaging techniques. However, they coincided in the specific morphology of the enhancement in only 11 nodules (52.4%): One nodule of 15 mm displayed homogeneous enhancement and 10 nodules showed peripheral enhancement in

6 HEPATOLOGY, Vol. 51, No. 6, 2010 VILANA ET AL Fig. 3. ICC in a 53-year-old man with cryptogenetic cirrhosis. (A) Baseline ultrasonography: 26 mm hypoechoic lesion in right hepatic lobe. (B) Arterial phase: peripheral (rim-like) enhancement. (C) Late phase: washout at 130 sec. (D,E) MRI of the same patient shows the nodule in the right hepatic lobe that presents a rim hyperenhancement that remains stable along the different arterial (D) and delayed (E) phases of the dynamic study. both techniques. On portal phase, no nodule was hyperenhanced on CEUS and, contrarily, all nodules displayed some kind of enhancement on MRI. Finally, on delayed phases all nodules displayed con- trast washout on CEUS; contrarily, on MRI all lesions remained hyperenhanced compared to surrounding parenchyma and in no cases was washout identified.

7 2026 VILANA ET AL. HEPATOLOGY, June 2010 Table 2. Ultrasonographic Appearance of ICC at Baseline and After Intravenous Contrast Enhancement in Arterial, Portal, and Delayed Phase Total (n521) 20 mm (n54) mm (n56) > 30 mm (n511) P Baseline pattern Hyperechoic, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Isoechoic, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Hypoechoic, n (%) 18 (85.7%) 4 (100%) 6 (100%) 8 (72.7%) Heterogeneous, n (%) 3 (14.3%) 0 (0%) 0 (0%) 3 (27.3%) Arterial phase Homogeneously hyperenhanced, n (%) 10 (47.6%) 3 (75%) 2 (33.3%) 5 (45.5%) Partially hyperenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Peripherally hyperenhanced, n (%) 11 (52.4%) 1 (25%) 4 (66.7%) 6 (54.5%) Isoenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Hypoenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Portal phase Homogeneously hyperenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Partially hyperenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Peripherally hyperenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Isoenhanced, n (%) 6 (28.6%) 3 (75%) 1 (16.7%) 2 (18.2%) Hypoenhanced, n (%) 15 (71.4%) 1 (25%) 5 (83.3%) 9 (81.8%) Delayed phase Homogeneously hyperenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Partially hyperenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Peripherally hyperenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Isoenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Hypoenhanced, n (%) 21 (100%) 4 (100%) 6 (100%) 11 (100%) Detection of washout <60 seconds 10 (47.6%) 1 (25%) 3 (50%) 6 (54.5%) seconds 5 (23.8%) 0 (0%) 2 (33.3%) 3 (27.3%) >120 seconds 6 (28.6%) 3 (75%) 1 (16.7%) 2 (18.2%) The analysis was done according to nodule size. Discussion The implementation of screening for HCC in cirrhosis patients by biannual abdominal US has led to the detection of small nodules. Once a nodule larger than 1 cm is detected a diagnostic work-up is recommended due to the high probability of malignancy. 7,8 Dynamic imaging techniques have a pivotal role in this setting and a noninvasive diagnostic criteria has been proposed and recently validated for HCC diagnosis taking into account a characteristic vascular pattern consisting of an intense contrast uptake during the arterial phase followed by washout in venous phases. 8,9 This vascular pattern has been shown to be specific for HCC diagnosis in several studies that have correlated the CT or MRI findings with explants. 21 CEUS has emerged as an instrumental tool for focal lesion characterization within a cirrhotic liver 22 and the American Association for the Study of Liver Diseases (AASLD) clinical practice guideline has incorporated CEUS, along with contrast-enhanced MRI imaging or CT scan, among the techniques that can be Fig. 4. Representation of the appearance of ICC nodules through the different phases by CEUS. Those nodules indistinguishable from HCC are highlighted. w/o, washout.

8 HEPATOLOGY, Vol. 51, No. 6, 2010 VILANA ET AL Table 3. MRI Characteristics of ICC at Baseline and After Intravenous Contrast Enhancement in Arterial, Portal, and Delayed Phase Total (n521) 20 mm (n54) mm (n56) > 30 mm (n511) P T1-weighted Hypointense, n (%) 18 (85.7%) 4 (100%) 5 (83.3%) 9 (81.8%) Isointense, n (%) 2 (9.5%) 0 (0%) 1 (16.7%) 1 (9.1%) Hyperintense, n(%) 1 (4.8%) 0 (0%) 0 (0%) 1 (9.1%) T2-weighted Hyperintense, n (%) 13 (61.9%) 2 (50%) 3 (50%) 8 (72.7%) Isointense, n (%) 4 (19.0%) 1 (25%) 2 (33.3%) 1 (9.1%) Heterogeneous, n (%) 2 (9.5%) 0 (0%) 0 (0%) 2 (18.2%) Hypointense, n (%) 2 (9.5%) 1 (25%) 1 (16.7%) 0 (0%) Arterial phase Homogeneously hyperenhanced, n (%) 2 (9.5%) 1 (25%) 1 (16.7%) 0 (0%) Partially hyperenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Peripherally hyperenhanced, n (%) 19 (90.5%) 3 (75%) 5 (83.3%) 11 (100%) Isoenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Hypoenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Portal phase Homogeneously hyperenhanced, n (%) 2 (9.5%) 1 (25%) 1 (16.7%) 0 (0%) Partially hyperenhanced, n (%) 9 (42.9%) 2 (50%) 1 (16.7%) 6 (54.5%) Peripherally hyperenhanced, n (%) 10 (47.6%) 1 (25%) 4 (66.7%) 5 (45.5%) Isoenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Hypoenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Delayed phase Homogeneously hyperenhanced, n (%) 11 (52.4%) 4 (100%) 3 (50%) 4 (36.4%) Partially hyperenhanced, n (%) 8 (38.1%) 0 (0%) 2 (33.3%) 6 (54.5%) Peripherally hyperenhanced, n (%) 2 (9.5%) 0 (0%) 1 (16.7%) 1 (9.1%) Isoenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Hypoenhanced, n (%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) The analysis was done according to nodule size. used for the noninvasive diagnosis of HCC in cirrhosis. 8 The incidence of ICC has increased in the last years and cirrhosis is a well-known risk factor for its development. 4,5 For that reason, this entity should be taken into consideration when a focal lesion is characterized, as the clinical management is sharply different and, thus, the need to distinguish these two entities is not trivial. 4 In that sense, liver transplantation is an established treatment option for HCC. Contrarily, ICC is not considered for liver transplantation in most centers. 4,23 The European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB) has recently published updated clinical practice recommendations for CEUS and they suggested that the typical enhanced pattern for cholangiocarcinoma is a rim-like enhancement (or nonenhancement) during the arterial phase followed by hypo/nonenhancement during the portal and delayed phases. 24 However, these recommendations are based on case-series of limited number of patients, most of them noncirrhosis, with large tumors. In addition, in a recent prospective study CEUS misdiagnosed a 16-mm ICC as HCC due to the detection of complete arterial contrast enhancement followed by washout. 9 This vascular profile was indistinguishable from HCC and also found in a recent series, 16 thus opening the debate of the real value of CEUS for distinguishing both entities. In our study, only 11 out of 21 pathologically confirmed ICC (52.4%) displayed the characteristic peripheral, rim-like enhancement that has been suggested as the most frequent finding in ICC ,24 This discrepancy may be due to the smaller nodule size in our patients (median size 32 mm) compared to other series. Larger tumors usually develop a central fibrotic area, justifying the peripheral, rim-like arterial contrast uptake in larger tumors. 25 This was demonstrated in our study, as more than half of nodules larger than 2 cm displayed a peripheral enhancement compared to only one nodule of the four nodules smaller than 2 cm. Furthermore, in a recent published article that evaluated the ICC appearance on CEUS, the nodules smaller than 3 cm showed more frequently a complete, homogeneous arterial contrast uptake compared to larger nodules, 16 and this finding was also demonstrated by dynamic CT scan. 26 In addition, 10 out of 21 cases (47.6%) displayed an intense and homogeneous contrast uptake during the arterial phase, followed by contrast washout on

9 2028 VILANA ET AL. HEPATOLOGY, June 2010 venous phases, the reported specific pattern described for HCC. Furthermore, seven of these 10 cases were nodules larger than 20 mm, so they had been misdiagnosed as HCC according to the noninvasive diagnostic criteria proposed by the AASLD. An experienced radiologist may argue that ICC may display, as some hypervascular metastasis, an intense arterial contrast uptake followed by a rapid contrast washout in the portal phase (<60 seconds). However, in our study five nodules (16, 18, 35, 35, and 70 mm) displayed a homogeneous arterial contrast uptake followed by contrast washout after 60 seconds, thus being fully indistinguishable from HCC. Therefore, strict application of the AASLD diagnostic recommendation would have led to a false HCC diagnosis in those three nodules larger than 20 mm if CEUS had been the only dynamic imaging technique applied and no further tumor staging through CT or MRI would be done. This would not confirm a specific pattern and this absence of concordant findings would raise some doubts about the diagnostic accuracy. 7,8 In all our patients an MRI scan was done for diagnosis and staging purposes. The most frequent finding was progressive contrast uptake through the different phases, and more important, in no cases was a washout detected, allowing a clear differentiation with HCC. 27 Progressive contrast enhancement along the different phases of the dynamic study may be also seen in hemangiomas. However, in this entity the contrast enhancement starts with a characteristic peripheral nodular enhancement, opposite the rim-like peripheral enhancement of ICC, which avoids its misdiagnosis as ICC. This dynamic contrast pattern sharply differs from the CEUS. This disagreement is due to the different pharmacokinetic behavior of both contrasts. Although US contrast agents are exclusively intravascular, the majority of currently approved contrast agents for CT and MRI are rapidly cleared from the blood pool into the extracellular space. The abundant fibrotic tissue that is typically observed in ICC is responsible for the persistent enhancement that is observed on delayed contrast-enhanced MR images, because the gadolinium chelate will accumulate in the interstitial space. Of course, this will not occur with US contrast agents that are pure intravascular tracers The results of this study raise major concerns about the use of CEUS as the only dynamic imaging technique for noninvasive diagnosis. To our knowledge, the only study that has prospectively evaluated the diagnostic accuracy of CEUS for HCC diagnosis showed that CEUS misclassified a 16-mm ICC as HCC. 9 However, this study only assessed newly detected, solitary nodules smaller than 2 cm and only one ICC was finally diagnosed, so this finding was considered anecdotal. This series showed that nearly 50% of the ICC lesions showed a homogeneous contrast uptake followed by washout. Therefore, the use of CEUS as the only imaging tool for noninvasive HCC diagnosis may be inappropriate and could not replace a dynamic MRI scan. CEUS may establish the malignant nature of a hepatic nodule in a cirrhotic liver, but should not pretend to establish its final diagnosis. If patients are not to be treated because of advanced disease, CEUS could be enough and no further imaging techniques should be necessary to implement best supportive care. However, if treatment is considered, staging will be needed, and at that point our data suggest that MRI will depict a pattern that is not HCC-specific and, hence, pathological confirmation should be requested. The limitation to our study is the relatively small number of patients reported. However, as we previously noted, to our knowledge this is the first study that analyzes the CEUS findings of ICC exclusively in cirrhosis patients, particularly when the disease is detected at earlier stages in the setting of screening. To conclude, ICC in cirrhotic liver may display a vascular pattern indistinguishable from HCC on CEUS. Taking into consideration that the need to distinguish these two entities is not trivial, because the different clinical management of both entities is sharply different, CEUS should not be used as the sole diagnostic tool for noninvasive diagnosis of HCC in cirrhosis and the current diagnostic recommendations proposed by the AASLD guidelines should be amended in that regard. References 1. de Groen PC, Gores GJ, LaRusso NF, Gunderson LL, Nagorney DM. Biliary tract cancers. N Engl J Med 1999;341: Khan SA, Thomas HC, Davidson BR, Taylor-Robinson SD. Cholangiocarcinoma. Lancet 2005;366: Shaib Y, El-Serag H. The epidemiology of cholangiocarcinoma. Semin Liver Dis 2004;24: Blechacz B, Gores GJ. Cholangiocarcinoma: advances in pathogenesis, diagnosis, and treatment. HEPATOLOGY 2008;48: Shaib YH, El-Serag HB, Nooka AK, Thomas M, Brown TD, Patt YZ, et al. Risk factors for intrahepatic and extrahepatic cholangiocarcinoma: a hospital-based case-control study. Am J Gastroenterol 2007;102: Organ distribution: allocation of livers liver transplant candidates with hepatocellular carcinoma (HCC). United Network for Organ Sharing Website. Available at: policies.asp?resourcestrue Revised March 3, Bruix J, Sherman M, Llovet JM, Beaugrand M, Lencioni R, Burroughs AK, et al. Clinical management of hepatocellular carcinoma.

10 HEPATOLOGY, Vol. 51, No. 6, 2010 VILANA ET AL Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver. J Hepatol 2001;35: Bruix J, Sherman M. Management of hepatocellular carcinoma. HEPA- TOLOGY 2005;42: Forner A, Vilana R, Ayuso C, Bianchi L, Sole M, Ayuso JR, et al. Diagnosis of hepatic nodules 20 mm or smaller in cirrhosis: Prospective validation of the noninvasive diagnostic criteria for hepatocellular carcinoma. HEPATOLOGY 2008;47: Chen LD, Xu HX, Xie XY, Lu MD, Xu ZF, Liu GJ, et al. Enhancement patterns of intrahepatic cholangiocarcinoma: comparison between contrast-enhanced ultrasound and contrast-enhanced CT. Br J Radiol 2008;81: D Onofrio M, Vecchiato F, Cantisani V, Barbi E, Passamonti M, Ricci P, et al. Intrahepatic peripheral cholangiocarcinoma (IPCC): comparison between perfusion ultrasound and CT imaging. Radiol Med 2008; 113: Li R, Guo Y, Hua X, He Y, Ding J, Guo A, et al. Characterization of focal liver lesions: comparison of pulse-inversion harmonic contrastenhanced sonography with contrast-enhanced CT. J Clin Ultrasound 200;35: Celli N, Gaiani S, Piscaglia F, Zironi G, Camaggi V, Leoni S, et al. Characterization of liver lesions by real-time contrast-enhanced ultrasonography. Eur J Gastroenterol Hepatol 2007;19: Xu HX, Liu GJ, Lu MD, Xie XY, Xu ZF, Zheng YL, et al. Characterization of focal liver lesions using contrast-enhanced sonography with a low mechanical index mode and a sulfur hexafluoride-filled microbubble contrast agent. J Clin Ultrasound 2006;34: Xu HX, Lu MD, Liu GJ, Xie XY, Xu ZF, Zheng YL, et al. Imaging of peripheral cholangiocarcinoma with low-mechanical index contrastenhanced sonography and SonoVue: initial experience. J Ultrasound Med 2006;25: Chen LD, Xu HX, Xie XY, Xie XH, Xu ZF, Liu GJ, et al. Intrahepatic cholangiocarcinoma and hepatocellular carcinoma: differential diagnosis with contrast-enhanced ultrasound. Eur Radiol 2010;20: Marti-Bonmati L, Talens A, del Olmo J, de Val A, Serra MA, Rodrigo JM, et al. Chronic hepatitis and cirrhosis: evaluation by means of MR imaging with histologic correlation. Radiology 1993;188: Schuppan D, Afdhal NH. Liver cirrhosis. Lancet 2008;371: Ito K, Mitchell DG, Hann HW, Kim Y, Fujita T, Okazaki H, et al. Viral-induced cirrhosis: grading of severity using MR imaging. AJR Am J Roentgenol 1999;173: Mortele KJ, Ros PR. MR imaging in chronic hepatitis and cirrhosis. Semin Ultrasound CT MR 2002;23: Burrel M, Llovet JM, Ayuso C, Iglesias C, Sala M, Miquel R, et al. MRI angiography is superior to helical CT for detection of HCC prior to liver transplantation: an explant correlation. HEPATOLOGY 2003;38: Lencioni R, Piscaglia F, Bolondi L. Contrast-enhanced ultrasound in the diagnosis of hepatocellular carcinoma. J Hepatol 2008;48: Khan SA, Davidson BR, Goldin R, Pereira SP, Rosenberg WM, Taylor- Robinson SD, et al. Guidelines for the diagnosis and treatment of cholangiocarcinoma: consensus document. Gut 2002;51(Suppl 6):VI1-VI Claudon M, Cosgrove D, Albrecht T, Bolondi L, Bosio M, Calliada F, et al. Guidelines and good clinical practice recommendations for contrast enhanced ultrasound (CEUS) update Ultraschall Med 2008;29: Chung YE, Kim MJ, Park YN, Choi JY, Pyo JY, Kim YC, et al. Varying appearances of cholangiocarcinoma: radiologic-pathologic correlation. Radiographics 2009;29: Kim SJ, Lee JM, Han JK, Kim KH, Lee JY, Choi BI. Peripheral massforming cholangiocarcinoma in cirrhotic liver. AJR Am J Roentgenol 2007;189: Rimola J, Forner A, Reig M, Vilana R, de Lope CR, Ayuso C, et al. Cholangiocarcinoma in cirrhosis: absence of contrast washout in delayed phases by magnetic resonance imaging avoids misdiagnosis of hepatocellular carcinoma. HEPATOLOGY 2009;50: Fan ZM, Yamashita Y, Harada M, Baba Y, Yamamoto H, Matsukawa T, et al. Intrahepatic cholangiocarcinoma: spin-echo and contrastenhanced dynamic MR imaging. AJR Am J Roentgenol 1993;161: Hamrick-Turner J, Abbitt PL, Ros PR. Intrahepatic cholangiocarcinoma: MR appearance. AJR Am J Roentgenol 1992;158: Gabata T, Matsui O, Kadoya M, Yoshikawa J, Ueda K, Kawamori Y, et al. Delayed MR imaging of the liver: correlation of delayed enhancement of hepatic tumors and pathologic appearance. Abdom Imaging 1998;23: Maetani Y, Itoh K, Watanabe C, Shibata T, Ametani F, Yamabe H, et al. MR imaging of intrahepatic cholangiocarcinoma with pathologic correlation. AJR Am J Roentgenol 2001;176: Wilson SR, Kim TK, Jang H-J, Burns PN. Enhancement patterns of focal liver masses: discordance between contrast-enhanced sonography and contrast-enhanced CT and MRI. AJR Am J Roentgenol 2007;189: W7-W12.