Hepatocellular Carcinoma Associated with Membranous Obstruction

Note: This copy is for your personal, non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights. ORIGINAL RESEARCH

Dong Il Gwon, MD Gi-Young Ko, MD Hyun-Ki Yoon, MD Kyu-Bo Sung, MD Jin Hyoung Kim, MD Seung Soo Lee, MD Jae Myeong Lee, MD Joon-Young Ohm, MD Ji Hoon Shin, MD Ho-Young Song, MD

1

From the Department of Radiology and Research Institute of Radiology, University of Ulsan, College of Medicine, Asan Medical Center, 388-1, Pungnap-2dong, Songpa-gu, Seoul 138-736, Korea. Received May 5, 2009; revision requested June 24; revision received July 1; accepted August 17; final version accepted September 2. Address correspondence to G.Y.K. (e-mail: [email protected]). q

Purpose:

To analyze the characteristics associated with membranous obstruction of the inferior vena cava (MOVC)– associated hepatocellular carcinoma (HCC) and to evaluate the clinical efficacy of transcatheter arterial chemoembolization (TACE).

Materials and Methods:

This retrospective study was approved by an institutional review board, and informed consent was waived. Ninetyeight patients (mean age, 48.5 years 6 12.9 [standard deviation]) with MOVC were retrospectively evaluated. The diagnosis of Budd-Chiari syndrome was confirmed with results from Doppler ultrasonography, computed tomography, magnetic resonance imaging, and/or inferior venacavography. The cumulative incidences of HCC and the patient survival period were calculated by using the Kaplan-Meier method. Factors associated with the development of HCC were evaluated by using multivariate Cox regression analysis.

Results:

Among 98 patients with MOVC, liver nodules were detected in 37 patients (38%), 23 of whom had HCC associated with MOVC and 14 of whom had benign nodules. The cumulative incidence of HCC at 1, 5, and 10 years was 7.3%, 13.5%, and 31.8%, respectively. Female sex was the only significant factor associated with the development of HCC (odds ratio, 6.02; P ,.001). HCC was of the single nodular type and of peripheral location. Among 23 patients with HCC, 20 patients were treated with only TACE and three with liver transplantation after TACE. After TACE, 14 (61%) of the study patients had a complete response, and survival rates at 1, 2, 3, 4, and 5 years were 90%, 85%, 61%, 61%, and 46%, respectively.

Conclusion:

The incidence of HCC in patients with MOVC was similar to that found in other studies. TACE resulted in an effective tumor response for HCC and seemed to be effective in prolonging patient survival. Female sex was the only significant factor associated with the development of HCC. A single nodular tumor with a peripheral location appears to have a higher probability of HCC. q

RSNA, 2010

RSNA, 2010

Radiology: Volume 254: Number 2—February 2010

n

radiology.rsna.org

617

n VASCULAR AND INTERVENTIONAL RADIOLOGY

Hepatocellular Carcinoma Associated with Membranous Obstruction of the Inferior Vena Cava: Incidence, Characteristics, and Risk Factors and Clinical Efficacy of TACE1

VASCULAR AND INTERVENTIONAL RADIOLOGY: Membranous Obstruction of Inferior Vena Cava

B

udd-Chiari syndrome (BCS) refers to a disorder associated with symptoms of portal hypertension caused by occlusion and/or stenosis of the hepatic veins or of the inferior vena cava in its hepatic portion. BCS consists of two forms—primary hepatic vein thrombosis and membranous obstruction of the inferior vena cava (MOVC). Patients with BCS more frequently have hepatocellular carcinoma (HCC) (1–3). The incidence of HCC in patients with MOVC varies in different studies and different world regions, as HCC risk is highly influenced by infection with the hepatitis B virus or the hepatitis C virus in highly endemic regions and by other causes of BCS (4–14). Therefore, in these studies, it is difficult to analyze the incidence of HCC, the risk factors for its development, and the radiologic and histologic characteristics of HCC in patients who only have MOVC. With regard to treatment, transcatheter arterial chemoembolization (TACE) has been increasingly accepted as an alternative therapeutic modality for unresectable HCC and has been shown to provide survival benefit in patients with unresectable HCC compared with those who were untreated (15,16). However, we have not found any study reporting

Advances in Knowledge n Female sex is the only significant factor associated with the development of hepatocellular carcinoma (HCC) in patients with membranous obstruction of the inferior vena cava (MOVC). n All MOVC-associated HCCs were revealed to be nodular and were located in the peripheral region of the liver, especially in the subcapsular or exophytic region. n Complete response after transcatheter arterial chemoembolization (TACE) was achieved in 61% of our study patients. n The 3- and 5-year survival rates were 64% and 50.4%, respectively, and were higher than those in previous studies.

618

the clinical efficacy of TACE or the longterm survival results in patients with MOVC-associated HCC. The purpose of this study was to determine the incidence of HCC in patients with isolated MOVC, to assess the potential risk factors of the development of HCC, to analyze the imaging characteristics of HCC, and to evaluate the efficacy of TACE and long-term survival results in patients with HCC associated with MOVC.

Materials and Methods Patient Population From May 1990 to November 2008, 159 consecutive patients with BCS diagnosed at the liver clinic of our institution were retrospectively studied. Demographic, clinical, and laboratory data were collected from the patients’ medical records or from the electronic patient information database. The protocol of this study was approved by the institutional review board of our institution, and informed consent was waived. Some of these patients were included in a previous publication (17). Figure 1 illustrates the flowchart of this study. The diagnosis of BCS was confirmed with results from Doppler ultrasonography (US), computed tomography (CT), magnetic resonance (MR) imaging, and/or inferior venacavography. Patients with tumorous hepatic vein or inferior vena cava invasion or compression (n = 23), those with hepatic vein thrombosis without MOVC (n = 17), and those with MOVC who had other risk factors such as a positive result for serum hepatitis B surface antigen or hepatitis C virus (n = 18) or alcohol consumption of more than 80 grams per day (n = 3) were excluded from the analysis. We therefore included 98 patients with only MOVC. The diagnosis Implications for Patient Care n Female patients with MOVC should be observed closely because they represent a highrisk group for HCC. n TACE appears to result in an effective treatment response in patients with MOVC-associated HCC.

Gwon et al

of MOVC was confirmed with the occlusion of the hepatic or suprahepatic segment of the inferior vena cava caused by a membrane with or without calcifications, and all patients with MOVC had imaging and clinical features of nonviral and nonalcoholic liver cirrhosis. CT was performed in all patients, inferior venacavography in 35 patients, MR imaging in 16 patients, and Doppler US in nine patients. The characteristics of the patients with MOVC are presented in Table 1. Thirty-seven of 98 patients with MOVC had liver nodules. Liver nodules were evaluated according to the following criteria: Multiple nodules (more than 10) with hypervascularity during the arterial phase and without washout during the portal venous phase and a serum AFP level less than 15 ng/dL during follow-up were diagnosed as benign lesions. A nodule with typical imaging features (hypervascular nodule with washout during the portal venous phase) and a serum AFP level greater than 400 ng/dL were the criteria for HCC. Biopsy was performed in nodules when these criteria were not met. Overall, liver nodules were diagnosed with percutaneous biopsy results in 12 patients and with imaging features and the serum AFP level in 25 patients. If a patient had more than one nodule, biopsy was performed in the dominant

Published online 10.1148/radiol.09090738 Radiology 2010; 254:617–626 Abbreviations: AFP = a-fetoprotein BCS = Budd-Chiari syndrome HCC = hepatocellular carcinoma MOVC = membranous obstruction of the inferior vena cava TACE = transcatheter arterial chemoembolization Author contributions: Guarantors of integrity of entire study, D.I.G., G.Y.K.; study concepts/study design or data acquisition or data analysis/ interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, D.I.G., J.H.K.; clinical studies, D.I.G., G.Y.K., H.K.Y., J.H.K., S.S.L., J.M.L., J.Y.O., J.H.S.; statistical analysis, D.I.G., J.H.K.; and manuscript editing, D.I.G., G.Y.K., H.K.Y., K.B.S., J.H.K., S.S.L., J.H.S., H.Y.S. Authors stated no financial relationship to disclose.

radiology.rsna.org

n



one. The following lesion criteria were evaluated in each patient: number, size of the largest lesion in each patient, location, and imaging features. Experienced radiologists (D.I.G., S.S.L.) each with at least 10 years of experience evaluated the lesion criteria and tumor response.

Figure 1

Figure 1: Flowchart of the review of these study patients. AFP = a-fetoprotein, LDLT = living donor liver transplantation.

Table 1 Characteristics of 98 Patients with MOVC Characteristic

Datum

Sex Male 48 (49) Female 50 (51) Age (y)* 48.5 6 12.9 Child-Pugh class A 73 (74) B 22 (22) C 3 (3) Treatment Conservative 60 (61) Angioplasty 20 (20) Inferior vena cava stent 7 (7) placement Transjugular intrahepatic 2 (2) portosystemic shunt Surgical portosystemic shunt 3 (3) Liver transplantation 6 (6) Note.—Unless otherwise indicated, data are numbers of patients, with percentages in parentheses. Percentages may not add up to 100% because of rounding. * Data are mean 6 standard deviation.


n

Imaging Contrast material–enhanced multiphase CT was performed by using a singledetector helical CT scanner (Somatom Plus; Siemens, Erlangen, Germany), a four-detector CT scanner (LightSpeed QX/I or LightSpeed Plus; GE Medical Systems, Milwaukee, Wis), or a 16detector CT scanner (Somatom Sensation 16; Siemens). A total of 120–150 mL of iopromide (Ultravist 300 or Ultravist 370; Schering, Berlin, Germany) was administered intravenously at a rate of 3 mL/ sec with an automatic power injector. CT images were obtained during the hepatic arterial phase (by using bolus-tracking methods or a 36-second delay), the portal venous phase (by using a 72-second delay), and the equilibrium phase (by using a 3-minute delay). Single-detector helical CT images were obtained with 7-mm collimation, a table speed of 10 mm per gantry rotation, 120 kVp, 200 mAs (effective), and 7-mm reconstruction intervals. Scanning parameters for fourdetector CT were as follows: 4 3 2.5-mm detector collimation, table speed of 15 mm per gantry rotation, 120 kVp, 200 mAs (effective), section thickness of 5 mm, and 5-mm reconstruction interval. Parameters for 16-detector CT were as follows: 16 3 1.5-mm detector collimation, table speed of 24 mm per gantry rotation, 120 kVp, 200 mAs (maximal effective) with the use of an automatic dose modulation technique (Care Dose; Siemens Medical Solutions, Forchheim, Germany), section thickness of 5 mm, and 5-mm reconstruction interval. Gadolinium-enhanced MR imaging was performed with a 1.5-T imager (Magnetom Avanto; Siemens) by using a six-element phased-array body coil as the receiver coil. MR imaging protocol consisted of T1-weighted breath-hold dual gradient-echo imaging, respiratorytriggered T2-weighted turbo spin-echo imaging, and gadolinium-enhanced coronal three-dimensional gradient-echo MR angiographic imaging with breath holding. Gadolinium-enhanced MR angiography was performed during the arterial phase (determined by using a real-time bolus display method [Care Bolus; Siemens]), the venous phase (50 seconds after contrast material administration),

radiology.rsna.org

Gwon et al

and the delayed phase (3 minutes after contrast material injection) after bolus injection of 30 mL of gadopentetate dimeglumine (Magnevist; Schering) at a rate of 2 mL/sec by using a power injector. Doppler US examinations were performed by experienced radiologists (D.I.G. and S.S.L., each with at least 10 years of experience) with a Sequoia 512 scanner (Acuson, Mountain View, Calif) with a 1–4-MHz convex probe, an HDI 5000 scanner (Philips Ultrasound, Bothell, Wash) with a 2–4-MHz convex probe, or a LOGIQ 700 scanner (GE Medical Systems) with a 2–4-MHz convex probe. Color and spectral Doppler sonograms were obtained in the inferior vena cava, hepatic veins, and portal veins. Inferior venacavograms were obtained by using a digital subtraction angiographic unit (V-3000, Philips Medical Systems, Best, the Netherlands; or Multistar TOP, Siemens Medical Solutions) and by using a 5-F catheter (Cobra; Cook, Bloomington, Ind) placed at the suprarenal inferior vena cava, and 30–40 mL of iopromide (Ultravist 300 or Ultravist 370; Schering) was administered at a rate of 6–8 mL/sec with a power injector (Angiomat 6000; LiebelFlarsheim, Cincinnati, Ohio).

TACE Procedure As the routine prior to TACE, superior mesenteric and celiac arteriography was initially performed to assess anatomy, tumor burden, vascularity, and portal venous patency. Depending on tumor burden, selective catheterization of segmental or subsegmental branches of the hepatic artery was then performed by using a 2-F microcatheter (SP; Terumo, Tokyo, Japan). Cisplatin (Cisplan; Dong-A, Seosan, Korea) was then infused for 15 minutes without embolic particle administration. The infused dose of cisplatin was 2 mg per kilogram of body weight. After cisplatin infusion, a mixture of iodized oil (Lipiodol; Laboratoire Guerbet, Aulnay-Sous-Bois, France) and cisplatin was infused into the feeding arteries until arterial flow stasis was achieved or iodized oil appeared in the portal branches. Embolization of the feeding arteries was performed by using 1-mm-diameter 619


absorbable gelfoam sponge particles (Gelfoam; Upjohn, Kalamazoo, Mich).

Follow-up In all patients with MOVC, follow-up physical examination, laboratory tests (blood count, liver function tests, and serum AFP level), and screening US of the liver were performed at 3–6-month intervals. In patients with HCC, follow-up physical examination, laboratory tests (blood count, liver function tests, and serum AFP level), and dynamic CT were performed after TACE on a routine basis at 1-month intervals. Tumor response was evaluated on the basis of CT images obtained 1–3 months (mean, 1.3 months 6 0.25 [standard deviation]) after TACE and was classified into four grades according to Response Evaluation Criteria in Solid Tumors guidelines as follows: complete response indicated complete iodized oil retention inside the tumor and subjective absence of arterial phase enhancement at CT; partial response, reduction of 30% or more in maximum tumor diameter; stable disease, neither partial response nor progressive disease; and progressive disease, increase of 20% or more in maximum tumor diameter (18–20). Tumor size was measured by using electronic calipers as the maximum diameter of the largest targeted tumor, and tumor response was evaluated by two clinically experienced radiologists (D.I.G., S.S.L.) by consensus. Repeated treatment was indicated when new or residual tumor was detected with typical imaging features of HCC and elevated AFP level, which met our diagnostic criteria of HCC. Treatment was terminated if a patient could not tolerate the procedure because of a decline in clinical status and laboratory values. Definitions Generally, HCC can develop everywhere in the liver. However, during the analysis of HCC associated with MOVC, we discovered typical location of the HCCs and we divided the location as follows: Peripheral location was defined as when a tumor was located in the peripheral third of the liver, and central location 620

was defined as when a tumor was located in the central two-thirds of the liver. Subcapsular location was defined as a tumor abutting the liver capsule or distortion of the liver capsule by less than 50% of the tumor volume. Exophytic location was defined as a tumor distorting the liver capsule by more than 50% of the tumor volume. Technical success was defined as successful catheter placement and targeted administration of chemotherapeutic agents.

Statistical Analysis The Wilcoxon signed rank test was used for preprocedural and postprocedural AFP level comparison. The cumulative incidence of HCC and the survival period after TACE were calculated according to the Kaplan-Meier method. Multivariate Cox regression analysis was used to evaluate the factors associated with the development of HCC in patients with MOVC. The following potential factors were included in this analysis: age, sex, inferior vena cava dilatation including angioplasty and stent placement, therapeutic portosystemic shunt, and portal vein thrombosis. Variables were selected in a stepwise forward selection manner. All statistical analysis was performed by using software (SPSS, version 14.0; SPSS, Chicago, Ill). A two-sided P value of less than .05 was considered to indicate a significant difference. Results Benign Nodules The incidence of benign nodules in patients with MOVC was 14% (14 of 98 patients). There were seven men and seven women, and mean age was 37 years 6 16.1 (age range, 18–63 years) at the time of diagnosis. Benign nodules were diagnosed at the time of initial diagnosis of MOVC in three patients, whereas they developed after a median period of 9.3 years (range, 0.6–34.1 years) in 11 patients. All of the benign nodules were always homogeneous hypervascular nodules during the arterial phase and were without washout during the portal venous phase, and the serum AFP levels were normal in

Gwon et al

all patients. The median diameter of the benign nodules was 2.3 cm 6 0.7 (range, 1.2–3.2 cm). Eleven of 14 patients with benign nodules had more than 10 lesions within the liver, while three patients each had a single lesion. In the three patients with a single lesion, the diagnosis was confirmed as a benign regenerative nodule at histologic examination. Two benign nodules were located in the central region of the liver, and the other benign nodule was located in the peripheral region of the liver. Only one patient with a histologic diagnosis of multiple benign regenerative nodules had an increase in the size and number of lesions 12 months after initial diagnosis; however, the imaging features and serum AFP level had not changed. After the 12-month followup, the nodules did not change in size or number for 3 years. In one patient with a single benign regenerative nodule, the nodule had disappeared by the 23-month follow-up. These two patients did not undergo any interventional or surgical treatment for MOVC. During the median follow-up of 68.8 months (range, 22–123 months), five patients died and nine patients were alive at the end of follow-up. The causes of deaths were hepatic failure (n = 3) and esophageal varix bleeding (n = 2).

HCC Incidence and Risk Factors In our study, the incidence of HCC in patients with MOVC was 23% (23 of 98 patients), and the number of HCCs was 33 in 23 patients. HCC was diagnosed at the time of the initial diagnosis of MOVC in six patients, whereas it developed after a median of 11.6 years (range, 2.7–26.4 years) in 17 patients. Mean patient age was 41.8 years 6 11.8 (age range, 15–55 years) at the diagnosis of MOVC and 48.7 years 6 11.6 (age range, 25–72 years) at the diagnosis of HCC. All patients with HCC had clinical features of liver cirrhosis, and histologic confirmation of HCC was obtained in eight patients with 10 lesions because of no washout during the delayed phase in four patients, normal AFP level in three patients, and the request of a clinician in one patient who met our diagnostic criteria of HCC.

radiology.rsna.org

n



According to the Kaplan-Meier analysis, the cumulative incidence of HCC from the time of the initial diagnosis of MOVC at 1, 5, and 10 years was 7.3%, 13.5%, and 31.8%, respectively, and the mean and median HCC development periods from the time of the initial diagnosis of MOVC were 207 (95% confidence interval: 172.5, 242.5) and 288 (95% confidence interval: 131.9, 444.1) months, respectively. In our multivariate Cox regression analysis, sex was the only significant factor associated with the development of HCC in patients with MOVC (odds ratio, 6.02; 95% confidence interval: 2.01, 18.04; P , .001). The cumulative incidence of HCC was significantly higher in female patients than male patients (P , .001) (Fig 2).

HCC Characteristics The patient characteristics of the 23 patients with MOVC-associated HCC are presented in Table 2. The serum AFP level was normal (,15 ng/mL) in three patients with HCC, whereas the median level was 672 ng/mL (interquartile range, 304–3280 ng/mL) in 20 patients. Six patients were treated with

portosystemic shunts prior to the development of HCC (angioplasty, n = 2; stent placement, n = 2; surgical portosystemic shunt, n = 2), with a median postoperative follow-up of 7.9 years (range, 4.7–10.0 years). The histologic findings of all eight HCC samples available for evaluation appeared to be well-differentiated without microscopic vascular and/or biliary invasion, and the surrounding liver histologic findings showed severe fibrosis or cirrhosis with venous congestion. Sixteen patients (70%) had a solitary nodule (Figs 3, 4), and seven patients (30%) had four or fewer nodules (Fig 5). The median diameter of the HCCs was 4.1 cm (range, 1–9 cm). Of the 32 HCCs in the 23 patients, 16 were located in the subcapsular region, 13 were considered exophytic, and three were located in the peripheral region. All HCCs were hypervascular, and 27 HCCs in 19 patients were heterogeneous, whereas five HCCs in four patients were homogeneous during the arterial phase. Washout during the portal venous phase was documented in 26 HCCs in 19 patients, whereas no washout was documented in six HCCs in

Gwon et al

four patients. Two (9%) of 23 patients with MOVC-associated HCC had portal vein tumor thrombus, and none of the 23 patients had radiologic evidence of biliary duct invasion.

Technical and Clinical Outcomes of TACE TACE was technically successful in all study patients. There was a median of five sessions per patient (range, 1–19 sessions). After TACE, 14 patients (61%) showed complete response (Figs 3–5), with a serial decrease observed at follow-up CT (Fig 3c). Four patients Table 2 Characteristics of 23 Patients with MOVC-associated HCC Characteristic

Datum

Male-to-female ratio 5:18 Age (y)* 48.7 6 11.6 (25–72) Child-Pugh class A 15 (65) B 7 (30) C 1 (4) AFP level (ng/mL)† 495 (62.3–2500) No. of nodules Single 16 (70) 2–4 7 (30) Median diameter (cm)* 4.1 6 2.4 (1–9) Lobar location Right lobe 11 (48) Left lobe 9 (39) Both 3 (13) Parenchymal location Central parenchyma 0 (0) Peripheral 23 (100) parenchyma Vascularity Hypervascularity 23 (100) Hypovascularity 0 (0) Enhancement pattern Homogeneous 4 (17) Heterogeneous 19 (83) Washout at delayed phase Washout 19 (83) No washout 4 (17) Portal vein thrombosis 2 (9)

Figure 2

Note.—Unless otherwise indicated, data are numbers of patients, with percentages in parentheses. Percentages may not add up to 100% because of rounding.

Figure 2: Kaplan-Meier curves show the incidence of HCC according to sex. Median incidence period was 120 months for female patients and 288 months for male patients (P , .001). Radiology: Volume 254: Number 2—February 2010

n

radiology.rsna.org

* Data are means 6 standard deviations, with ranges in parentheses. †

Data in parentheses are the range.

621


Figure 3

Figure 3: Images in 55-year-old woman with MOVC-associated HCC who underwent one session of TACE. (a) Axial arterial phase dynamic CT image shows exophytic hypervascular HCC (5 3 4 cm) (arrows) at segment II. (b) Common hepatic angiogram shows hypervascular mass (arrows) supplied by segment II of the hepatic artery. (c) Axial arterial phase dynamic CT image obtained 4 years after TACE shows complete iodized oil retention inside the HCC (arrows) and a decrease in tumor size with no recurrence.

Figure 4

Figure 4: Data in 48-year-old woman with MOVCassociated HCC who underwent liver transplantation after one session of TACE. (a) Coronal portal venous phase dynamic CT image shows exophytic HCC (5 3 5 cm) (arrows) at segment II. Selective TACE was performed through segment II of the hepatic artery (not shown). (b) Coronal arterial phase dynamic CT image obtained 1 month after TACE shows complete iodized oil retention inside the HCC (arrows) without recurrence and a decrease in tumor size. (c) Photograph shows the cut surface of the well-defined HCC (arrows) is yellowish tan and granular with complete necrosis.

(17%) showed partial response. One patient showed no radiologic response to the procedure, and four patients showed 622

progression of HCC despite repeated TACE. The median pre-TACE AFP level was 495 mg/dL (interquartile range,

Gwon et al

62.3–2500 mg/dL), and the median post-TACE AFP level was 66.1 mg/dL (interquartile range, 8.3–237 mg/dL). A significant reduction of AFP levels after TACE was observed (P , .001). Seven (30%) of 23 patients had a recurrence after the initial diagnosis of HCC (median, 15.7 months; range, 6–32 months). All of the recurrent HCCs showed typical imaging features of HCC, and the serum AFP level was increased (median, 569 mg/dL; interquartile range, 450–744 mg/dL). A significant increase of AFP level was observed after recurrence (P = .018). In seven patients with recurrent HCCs, three patients had multiple HCCs, one patient had three HCCs, and three patients each had only one HCC. The size of all recurrent HCCs was less than 2 cm (mean, 1.6 cm; range, 1–2 cm). In four patients with one to three recurrent HCCs, all HCCs were located in the peripheral region of the other segments, whereas in three patients with multiple recurrent HCCs, all of the HCCs were located in both the central and peripheral regions of the same and other segments. Five (22%) of 23 patients had extrahepatic metastases after a median of 31.2 months (range, 20–42 months), with lung metastasis being the most common followed by bone metastasis. Nine patients (39%) experienced mild postembolization syndrome after undergoing TACE, including nausea, vomiting, right upper quadrant pain, and fever, but those symptoms disappeared within 7 days. One patient experienced inferior vena cava thrombosis after undergoing a second session of TACE, but it disappeared after 3 months of anticoagulation therapy.

TACE Survival Analysis During the median follow-up of 49.7 months (range, 5–114.5 months), 12 patients died and 11 patients were alive. The causes of death were pneumonia and sepsis (n = 5), hepatic failure (n = 4), and esophageal varix bleeding (n = 3). In the 20 patients who underwent only TACE, according to Kaplan-Meier analysis, the mean and median survival periods from initial diagnosis of HCC

radiology.rsna.org

n



Gwon et al

Figure 5

Figure 5: Images in 49-year-old woman with four HCCs associated with MOVC. (a, b) Axial arterial phase dynamic CT images show hypervascular HCCs located (a) in the subcapsular region (white arrow) of segment VII and (b) in the exophytic region (black arrow) of segment III. Two other HCCs were located in the subcapsular region of segment VI (not shown). (c) Common hepatic angiogram shows four hypervascular masses. White arrow indicates HCC in segment VII. Black arrow indicates HCC in segment III. Arrowheads indicate HCCs in segment VI. (d) Common hepatic angiogram from selective TACE performed through each segmental artery. (e, f) Axial arterial phase dynamic CT images obtained 2 months after TACE show complete iodized oil retention inside the HCCs (white and black arrow) with no recurrence and a decrease in tumor size.

were 65.8 (95% confidence interval: 46.9, 84.7) and 61.6 (95% confidence interval: 15.6, 107.6) months, respectively. The survival rates from initial diagnosis of HCC at 1, 2, 3, 4, and 5 years were 90%, 85%, 61%, 61%, and 46%, respectively (Fig 6). Three patients underwent living donor liver transplantation, and the mean period between TACE and living donor liver transplantation was 3 months (range, 2–5 months). Only one TACE session was performed in two patients, and three TACE sessions were performed in the other patient. Total necrosis of three HCCs in these three patients was confirmed with histologic examination results after living donor liver transplantation. During the median follow-up period after living donor liver transplantation (22.9 months; range, 13–32.8 months), these three patients remained alive and without tumor recurrence. Radiology: Volume 254: Number 2—February 2010

n

radiology.rsna.org

Figure 6

Figure 6: Cumulative patient survival rates according to Kaplan-Meier analysis.

623


Discussion Liver regenerative nodules are frequently encountered in patients with BCS and are considered a response to focal loss of portal perfusion and hyperarterialization in areas with preserved hepatic venous outflow (21–23). In this study, the incidence of benign nodules was 14%, and they developed within a median of 9.3 years (range, 0.6–34.1 years) after the initial diagnosis of MOVC in 11 of 14 patients. All 14 patients with benign nodules had normal serum AFP levels and typical imaging features. Vilgrain et al (22) reported that multiple (.10) small lesions were suggestive of benignity. In our study, 11 (79%) of 14 patients with benign nodules had multiple small nodules, whereas three (21%) patients had single small nodules. Histologic confirmation of a single nodule in the three patients was possible, and all nodules were identified as benign regenerative nodules. The incidence of HCC in patients with MOVC varies greatly according to world region and study, that is, 40.0% (6) and 47.5% (7) in South Africa, 6.4% (9) and 41% (10) in Japan, 11% (11) in India, and 4.7% (12) in Nepal. Variable rates of HCC development may be caused by differences in the patient population in terms of follow-up period, severity of cirrhosis at the time of diagnosis, environmental carcinogens, and inclusion of patients with chronic viral hepatitis (4–14). This study was a large retrospective study including 98 patients with MOVC and excluding patients with other causes of BCS as well as those who had MOVC concomitant with viral infection or chronic alcohol consumption. Therefore, the results of our study were caused only by MOVC. The incidence of HCC was 23%, similar to that reported for the previously mentioned studies, and the 5-year cumulative incidence was 13.5%, which was also similar to that reported for other chronic liver diseases (24). Moucari et al (14) reported that inferior vena cava obstruction was a major predictor for HCC development in patients with BCS; however, the cause of HCC development in patients with 624

MOVC is still unclear. Previous studies (13,25) have shown that hepatic congestion over a period of several years was one possible factor involved in MOVCassociated HCC and that liver fibrosis as well as cirrhosis may also be factors involved in MOVC-associated HCC. In our study, six patients had a diagnosis of HCC at the time of initial diagnosis of MOVC, and 17 patients had a chronic course of MOVC (median, 11.6 years; range, 2.7–26.4 years) with imaging features of cirrhosis and pathologic findings of severe fibrosis or cirrhosis. Therefore, long-standing MOVC may be strongly associated with the development of HCC in a fibrotic or cirrhotic context. As it is widely accepted that chronic infection by hepatitis B virus or hepatitis C virus can induce HCC as a result of chronic liver injury, the question is then whether hepatitis viral infection has any role in HCC development in patients with MOVC. Previous studies (9,26) have shown that hepatitis B virus infection was one possible factor associated with hepatocarcinogenesis in patients with MOVC. However, Simson (7) reported that the positive rate of hepatitis B surface antigen was low in patients with MOVC and HCC and speculated that the hepatitis B virus was not significantly involved in hepatocarcinogenesis. As for the hepatitis C virus, its role in MOVC-associated hepatocarcinogenesis will require further study. The reported characteristics of patients with BCS associated with MOVC are that HCC develops at a younger age and has a poorer prognosis compared with that of patients with HCC and no MOVC (6,9). However, in our study, MOVC-associated HCC developed in middle-aged patients (mean, 48.7 years; age range, 25–72 years), and those with MOVC-associated HCC had a good prognosis due to the effectiveness of TACE. It could therefore be speculated that hepatitis viral infection might have a role in HCC development and might influence patient prognosis. Shin et al (17) reported that patients with BCS-associated HCC (MOVC was 53%) might have greater benefit from being treated with aggressive intervention such as TACE or surgical resection

Gwon et al

than patients with hepatitis B virus– associated HCC because of their better hepatocellular function and the lower degree of invasiveness of the associated HCC. In this study, although there was a relatively small number of HCC cases, HCC developed more frequently in female patients, and the female sex was a single independent factor for the development of HCC associated with MOVC. Shin et al (17) postulated that estrogen may enhance the development of HCC in patients with BCS. Takamura et al (27) reported that even after successful treatment for the relief of congestion patients with MOVC should be followed closely for as long as possible because of the risk of HCC development. In our study, treatment of MOVC was not an independent factor for the development of HCC. Therefore, we are in agreement that despite portosystemic shunt treatment of MOVC, patients with MOVC have a risk of not only the occurrence but also the recurrence of HCC. All MOVC-associated HCCs were revealed to be nodular and were located in the peripheral region of the liver, especially in the subcapsular or exophytic region. It could therefore be speculated that the fibrotic process following chronic liver injuries might contribute mainly to the development of the nodular type of HCC which might usually initiate in the peripheral region of the liver. Furthermore, the peripheral location of the tumor seems to be one of the characteristic features of the tumor. In our study, two (9%) patients with MOVC-associated HCC had portal vein invasion at the time of diagnosis, and none of our patients had biliary duct invasion. The lower invasiveness of HCC associated with MOVC can be explained by the characteristics of the tumor as nodular and well-differentiated HCC, as well as that extensive hepatic fibrosis inhibits the invasion of tumor cells into the vascular or biliary tree. Moucari et al (14) reported that serum APF levels seemed to be a useful screening tool for patients with BCS. In our study, the median serum AFP level was 672 ng/mL (interquartile range,

radiology.rsna.org

n



304–3280 ng/mL) in 87% of patients with MOVC-associated HCC. In contrast, the median serum AFP level was 2.75 ng/dL (interquartile range, 2.0– 4.3 ng/dL) in all patients with MOVCassociated benign nodules. Therefore, serial follow-up of the serum AFP level could be useful in screening for the development of HCC in patients with MOVC. Moreover, AFP levels seem to be helpful to determine TACE response, as levels were significantly reduced after TACE. They appear to be useful to evaluate recurrence after TACE, as the serum AFP levels were significantly increased after recurrence. According to the European Association for the Study of the Liver guidelines for the detection of HCC, the suggested interval for surveillance in patients with cirrhosis has been set at 6 months (28). Because the incidence and the cumulative incidence of HCC in patients with MOVC were similar to that reported for other chronic liver diseases, the follow-up interval can be set at 6 months. Because female sex is a high-risk group for HCC, we recommend strict adherence to 6-month follow-up or perhaps even a shorter follow-up interval. In general, the prognosis in patients with HCC still remains poor because of the advanced stage of their cancers and the associated hepatic impairment seen at the time of diagnosis and because of the high intrahepatic recurrence rate of 79%–80% at 5 years after hepatic surgery, which results from either intrahepatic metastases from the primary tumor or multicentric occurrence (29,30). Tumor burden, hepatic functional reserve, performance status, tumor characteristics (ie, size, number, and presence of portal vein tumor thrombus; AFP level), and response to treatment have been noted as prognostic factors in general (31–35). In previous studies (36,37), TACE resulted in a partial response or better in 15%–55% of patients with HCC. Complete response to TACE was achieved in 61% (14 of 23) of our study patients. Therefore, in our study, tumor response to TACE in patients with MOVC-associated HCC was more favorable than that in the previously mentioned studies. AlRadiology: Volume 254: Number 2—February 2010

n

though histologic findings were not obtained in all our patients, we suggest that these good responses might be because of the level of differentiation and low invasiveness of the HCC we encountered. In this study, recurrent HCCs were observed in seven (30%) of 23 patients at a median of 15.7 months after the initial diagnosis. Most of the recurrent HCCs were located in the peripheral parenchyma of the other segments of the liver. Although histologic confirmation was not obtained in all our patients, we considered these HCCs to represent multicentric occurrence rather than intrahepatic metastasis, because most of the initial HCCs were well-differentiated masses without vascular invasion and most of the recurrent HCCs developed in other segments of the liver. Previous studies in patients with HCC have reported that TACE had 3- and 5-year survival rates of 26%–47% (15,16,37) and 16%–26% (38–40), respectively. In our study, the 3- and 5-year survival rates were 64% and 50.4%, respectively, and were higher than those in the previously mentioned studies. Because of the relatively low number of TACE cases, prognostic factor assessment with a multivariate analysis has not been attempted. Although the reason for better survival in our results are not known, it may be because most of our patients with MOVC-associated HCC had good liver function as well as good tumor response to TACE. Our study had several limitations. First, this study was retrospective over a long period of time (18 years), and many improvements in CT techniques for assessment of tumor vascularity have occurred. However, we think lesion vascularization was adequately assessed in our study population. Second, the diagnosis and histologic feature of well-differentiated HCC were confirmed in only 35% of patients with HCC. Furthermore, in half of the patients with HCC proved by means of histologic examination, specimens were obtained only from the largest nodules; therefore, the diagnosis of the other lesions was not confirmed. Third, because HCCs were detected during the regular

radiology.rsna.org

Gwon et al

follow-up period in 74% of patients with MOVC-associated HCC, perhaps earlier detection could have been possible and survival periods (lead-time bias) could have been increased. Finally, because this was a retrospective study, we have no control group to which to compare the response of our study population to TACE, but only historic control subjects, which seem to indicate TACE is advantageous in patients with MOVCassociated HCC. In conclusion, the incidence and the cumulative incidence of MOVC-associated HCC in our study were similar to those of other reports. Female sex was the only significant factor associated with the development of HCC in patients with MOVC. A single nodular lesion with a peripheral location appeared to have a higher probability for the diagnosis of HCC. Furthermore, TACE appeared to result in effective treatment response.

References 1. Okuda K, Kage M, Shrestha SM. Proposal of a new nomenclature for Budd-Chiari syndrome: hepatic vein thrombosis versus thrombosis of the inferior vena cava at its hepatic portion. Hepatology 1998;28:1191–1198. 2. Janssen HL, Garcia-Pagan JC, Elias E, et al. Budd-Chiari syndrome: a review by an expert panel. J Hepatol 2003;38:364–371. 3. Valla DC. The diagnosis and management of the Budd-Chiari syndrome: consensus and controversies. Hepatology 2003;38:793–803. 4. Takeuchi J, Takada A, Hasumura Y, Matsuda Y, Ikegami F. Budd-Chiari syndrome associated with obstruction of the inferior vena cava. Am J Med 1971;51:11–20. 5. Espana P, Figuera D, Femadez de Miguel JM, Anaya A, Menendez J, Durantez A. Membranous obstruction of the inferior vena cava and hepatic veins. Budd-Chiari syndrome?—a treatable disease. Am J Gastroenterol 1980;73:28–32. 6. Kew MC, McKnight A, Hodkinson J, Bukofzer S, Esser JD. The role of membranous obstruction of the inferior vena cava in the etiology of hepatocellular carcinoma in southern African blacks. Hepatology 1989;9:121–125. 7. Simson IW. Membranous obstruction of the inferior vena cava and hepatocellular

625


carcinoma in South Africa. Gastroenterology 1982;82:171–178. 8. Kage M, Arakawa M, Kojiro M, Okuda K. Histopathology of membranous obstruction of the inferior vena cava in the BuddChiari syndrome. Gastroenterology 1992; 102:2081–2090. 9. Okuda H, Yamagata H, Obata H, et al. Epidemiological and clinical features of Budd-Chiari syndrome in Japan. J Hepatol 1995;22:1–9. 10. Nakamura T, Nakamura S, Aikawa T, Suzuki O, Onodera A, Karoji N. Obstruction of the inferior vena cava in the hepatic portion and hepatic vein: report of 18 cases and review of the Japanese literature. Angiology 1968;19:479–498. 11. Dilawari JB, Mambery P, Chawla Y, et al. Hepatic venous flow obstruction (BuddChiari syndrome): experience with 177 patients and a review of the literature. Medicine (Baltimore) 1994;73:21–36. 12. Shrestha SM, Okuda K, Uchida T, et al. Epidemicity and clinical picture of liver disease due to obstruction of the hepatic portion of the inferior vena cava in Nepal. J Gastroenterol Hepatol 1996;11:170–179. 13. Matsui S, Ichida T, Watanabe M, et al. Clinical features and etiology of hepatocellular carcinoma arising in patients with membranous obstruction of the inferior vena cava: in reference to hepatitis viral infection. J Gastroenterol Hepatol 2000;15:1205–1211. 14. Moucari R, Rautou PE, Cazals-Hatem D, et al. Hepatocellular carcinoma in BuddChiari syndrome: characteristics and risk factors. Gut 2008;57:828–835. 15. Llovet JM, Real MI, Montana X, et al. Arterial embolisation or chomoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomized controlled trial. Lancet 2002; 359:1734–1739. 16. Lo CM, Ngan H, Tso WK, et al. Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma. Hepatology 2002; 35:1164–1171. 17. Shin SH, Chung YH, Suh DD, et al. Characteristic clinical features of hepatocellular carcinoma associated with Budd-Chiari syndrome: evidence of different carcinogenic process from hepatitis B virus-associated hepatocellular carcinoma. Eur J Gastroenterol Hepatol 2004;16:319–324.

626

18. Park JO, Lee SI, Song SY, et al. Measuring response in solid tumors: comparision of RECIST and WHO response criteria. Jpn J Clin Oncol 2003;33:533–537. 19. Georgiades CS, Hong K, D’Angelo M, Geschwind JF. Safety and efficacy of transarterial chemoembolization in patients with unresectable hepatocellular carcinoma and portal vein thrombosis. J Vasc Interv Radiol 2005;16:1653–1659. 20. Brown DB, Gould JE, Gervais DA, et al. Transcatheter therapy for hepatic malignancy: standardization of terminology and reporting criteria. J Vasc Interv Radiol 2007;18:1469–1478. 21. Wanless IR. Regenerative nodules in Budd-Chiari syndrome. Hepatology 1994; 19:1391. 22. Vilgrain V, Lewin M, Vons C, et al. Hepatic nodules in Budd-Chiari syndrome: imaging features. Radiology 1999;210:443–450. 23. Tanaka M, Wanless IR. Pathology of the liver in Budd-Chiari syndrome: portal vein thrombosis and the histogenesis of venocentric cirrhosis, veno-portal cirrhosis, and large regenerative nodules. Hepatology 1998; 27:488–496. 24. Fattovich G, Stroffolini T, Zagni I, Donato F. Hepatocellular carcinoma in cirrhosis: incidence and risk factors. Gastroenterology 2004;127:S35–S50. 25. Kawaguchi T, Sata M, Ono N, et al. BuddChiari syndrome complicated by hepatocellular carcinoma with no evidence of infection with hepatitis virus: a case report. Hepatogastroenterology 1999;46:3237–3240. 26. Takayasu K, Muranatsu Y, Moriyama N, et al. Radiological study of idiopathic BuddChiari syndrome complicated by hepatocellular carcinoma: a report of four cases. Am J Gastroenterol 1994;89:249–253. 27. Takamura M, Ichida T, Yokoyama J, Matsuda Y, Nomoto M, Aoyagi Y. Recurrence of hepatocellular carcinoma 102 months after successful eradication and removal of membranous obstruction of the inferior vena cava. J Gastroenterol 2004;39:681–684. 28. Bruix J, Sherman M, Llovet JM, et al. Clinical management of hepatocellular carcinoma: conclusions of the Barcelona-2000 EASL conference. J Hepatol 2001;35:421– 430. 29. Chen MF, Hwang TL, Jeng LB, Wang CS, Jan YY, Chen SC. Postoperative recurrence of hepatocellular carcinoma: two hundred

Gwon et al

five consecutive patients who underwent hepatic resection in 15 years. Arch Surg 1994;129:738–742. 30. Imamura H, Matsuyama Y, Tanaka E, et al. Risk factors contributing to early and late phase intrahepatic recurrence of hepatocellular carcinoma after hepatectomy. J Hepatol 2003;38:200–207. 31. Okuda K, Ohtsuki T, Obata H, et al. Natural history of hepatocellular carcinoma and prognosis in relation to treatment: study of 850 patients. Cancer 1985;56:918–928. 32. Llovet JM, Bru C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 1999;19: 329–338. 33. Prospective validation of the CLIP score: a new prognostic system for patients with cirrhosis and hepatocellular carcinoma. The Cancer of the Liver Italian Program (CLIP) Investigators. Hepatology 2000;31:840–845. 34. Kudo M, Chung H, Haji S, et al. Validation of a new prognostic staging system for hepatocellular carcinoma: the JIS score compared with the CLIP score. Hepatology 2004;40:1396–1405. 35. Park KW, Park JW, Choi JI, et al. Survival analysis of 904 patients with hepatocellular carcinoma in a hepatitis B virusendemic area. J Gastroenterol Hepatol 2008;23:467–473. 36. Ebied OM, Federle MP, Carr BI, et al. Evaluation of responses to chemoembolization in patients with unresectable hepatocellular carcinoma. Cancer 2003;97:1042–1050. 37. Bruix J, Sala M, Llovet JM. Chemoembolization for hepatocellular carcinoma. Gastroenterology 2004;127:S179–S188. 38. Takayasu K, Arii S, Ikai I, et al. Prospective cohort study of transarterial chemoembolization for unresectable hepatocellular carcinoma in 8510 patients. Gastroenterology 2006;131:461–469. 39. Hatanaka Y, Yamashita Y, Takahashi M, et al. Unresectable hepatocellular carcinoma: analysis of prognostic factors in transcatheter management. Radiology 1995; 195:747–752. 40. Savastano S, Miotto D, Casarrubea G, Teso S, Chiesura-Corona M, Feltrin GP. Transcatheter arterial chemoembolization for hepatocellular carcinoma in patients with Child’s grade A or B cirrhosis: a multivariate analysis of prognostic factors. J Clin Gastroenterol 1999;28:334–340.

radiology.rsna.org

n
