Venous Thrombosis After Radiofrequency Ablation for Hepatocellular Carcinoma

Transcription

1 Vascular and Interventional Radiology Original Research Kim et al. Venous Thrombosis After Radiofrequency Ablation Vascular and Interventional Radiology Original Research Ah Yeong Kim 1 Hyunchul Rhim Minjung Park Min Woo Lee Young-Sun Kim Dongil Choi Hyo K. Lim Kim AY, Rhim H, Park M, et al. Keywords: complication, hepatocellular carcinoma, radiofrequency ablation, venous thrombosis DOI: /AJR Received December 29, 2010; accepted after revision April 17, All authors: Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunwan University School of Medicine, 50 Ilwon-dong, Kangnam-ku, Seoul , Republic of Korea. Address correspondence to H. Rhim (rhimhc@skku.edu). AJR 2011; 197: X/11/ American Roentgen Ray Society Venous Thrombosis After Radiofrequency Ablation for Hepatocellular Carcinoma OBJECTIVE. This study was designed to evaluate the frequency, morphological patterns, sequential changes, and clinical significance of venous thrombosis after radiofrequency ablation (RFA) for hepatocellular carcinoma (HCC). MATERIALS AND METHODS. A total of 1379 RFAs performed in 1046 patients with HCC (mean tumor size, 1.93 cm) were surveyed. We retrospectively reviewed all radiologic reports before and after RFA and selected 15 patients with newly developed procedurerelated venous thrombosis. Procedure-related thrombosis was defined as being adjacent to the ablation zone and developing within 4 months after the procedure. We evaluated the frequency, morphological patterns, sequential changes, and clinical course of venous thrombosis (mean follow-up, days). Four cases with local tumor progression were identified among the 15 patients, and their clinical implications were evaluated. RESULTS. A total of 15 venous thromboses (1.08%; 12 portal and three hepatic veins) developed after RFA (range, days; mean, 37 days). The thromboses were found in central (n = 10), peripheral (n = 4), and both central and peripheral (n = 1) locations in the ablation zones. Thrombosis was decreased in nine (69.2%), persisted in one (7.6%), and increased in three (23.0%) of 13 patients who underwent follow-up CT for more than 12 months. Local tumor progression was noted in four patients (26.6%); it abutted to venous thrombosis in two patients, separated from the venous thrombosis in one patient, and exhibited malignant thrombosis in one patient. CONCLUSION. The development of portal or hepatic venous thromboses after RFA in patients with HCC is rare and usually is associated with favorable prognoses. Further investigation is warranted to elucidate whether venous thrombosis after RFA is related to local tumor progression around ablation zones. H epatocellular carcinoma (HCC) is one of the most common tumors worldwide. Surgical resection is the treatment of choice for HCCs in noncirrhotic patients [1]. However, because of the limited inclusion criteria for surgery, only 5% of patients in Western countries and 40% in Asia are suitable candidates for surgical resection [1, 2]. Consequently, several nonsurgical therapies, including radiofrequency ablation (RFA) [2 6], percutaneous ethanol injection [7], microwave coagulation therapy [8, 9], cryoablation therapy [10], transarterial embolization, and transarterial chemoembolization (TACE) [11, 12], have been developed. Among these therapies, RFA has been accepted as a promising technique for treating unresectable HCCs and has been increasingly used because of its good therapeutic efficacy and acceptable safety [1, 4, 13, 14]. According to the American Association for the Study of Liver Diseases practice guidelines, RFA provides equivalent or better local disease control compared with percutaneous ethanol injection in small HCC tumors [1]. However, a broad spectrum of complications, including intraperitoneal hemorrhage, liver abscess, hemobilia, biliary strictures, pneumothorax, pleural effusion, injury to adjacent organs, and liver failure, have been reported [4, 15, 16] with a prevalence of % [16, 17]. Among treatment-related complications after RFA, portal or hepatic venous thrombosis can occur as a result of thermal injury to the adjacent vessels during the ablation [15, 16]. Although portal and hepatic venous thrombosis are known as minor complications after RFA [5, 15, 16], there have been only a few studies on the patterns or clinical importance of venous thrombosis AJR:197, December 2011

2 Venous Thrombosis After Radiofrequency Ablation In this study, we evaluate the frequency, morphologic patterns, and sequential changes of portal or hepatic vein thrombosis after percutaneous RFA for HCC, especially focusing on the clinical outcome and relationship to local tumor recurrence, in a large singleinstitution series. Materials and Methods Subjects Between February 2005 and October 2009, a total of 1046 patients (812 men and 234 women) with HCCs (1379 tumors) underwent RFA at our institution and were surveyed for this study. Consent to inclusion was waived under the approval of the institutional review board. All patients who were treated with RFA for HCCs were followed up with contrast-enhanced three phase CT immediately and 1 month after RFA. The images were obtained before and at 30, 60 70, and 180 seconds after IV injection of contrast material. Complete tumor necrosis (successful treatment) is considered to have occurred when follow-up CT reveals no foci of enhancement within the ablated lesion or at its periphery [18]. If the RFA achieved a technical success at this time, follow-up CT was repeated every 3 months after this visit. In this retrospective study, we reviewed all of the radiologic reports of CT or MRI obtained after RFA and identified 103 patients who presented with portal or hepatic venous thrombosis. Radiologic reports were searched using the radiologic information system by inserting keywords, such as venous thrombosis, for patients who underwent RFA for HCCs. Among these 103 patients, 88 had portal or hepatic venous thrombosis that was not considered to be related to RFA according to the radiologic report of liver CT or MRI after the procedure. Venous thrombosis was defined by the presence of a nonenhancing filling defect within the portal or hepatic vein. In this study, we defined ablation-related venous thromboses as those within 2 cm from the ablation margin resulting from thermal damage, which is presumed to affect adjacent vessels. Thus, 88 of 1046 patients (8.41%) who had venous thrombosis distal to 2 cm from the ablated margin were excluded from this study. Thromboses that were detected by CT scan within 4 months after RFA were regarded as consequences of the procedure. Fifteen patients with portal or hepatic venous thrombosis as a consequence of a previously performed RFA were selected from the 1046-patient cohort for inclusion in this study and were followed up (mean follow-up, days; range, days). Among the 15 patients, one (patient 11) underwent a 4-month follow-up CT scan at 4 days, postponed regardless of medical condition, and was included in this cohort. This cohort consisted of 13 men and two women (age range, years; mean age, 58.8 years). Eleven patients (73%) had liver cirrhosis as a result of hepatitis B (n = 8), hepatitis C (n = 2), or alcoholism (n = 1). Four patients had chronic hepatitis B without cirrhosis. Tumor size ranged from 1.1 to 4.5 cm in maximal diameter (mean size, 2.5 cm). The diagnosis of HCC was confirmed by percutaneous needle biopsy in one tumor, and the remaining 14 tumors were considered to be HCCs by characteristic CT or MRI findings and elevated serum tumor markers (α-fetoprotein level, > 400 μg/l) on the basis of Barcelona Clinic Liver Cancer and American Association for the Study of Liver Diseases diagnosis criteria. Thirteen RFA procedures were performed under real-time ultrasound guidance, whereas the remaining two procedures were performed intraoperatively. For all 15 patients, demographic information, cause of liver disease, size and location of tumors that were ablated, and type of ablative technique were recorded (Table 1). Diagnosis and Treatment Guidelines of HCCs Diagnosis and treatment of HCCs in this study followed the Barcelona Clinic Liver Cancer staging system and American Association for the Study of Liver Disease [1] diagnosis and treatment guidelines. In this study, diagnosis of HCC was made by percutaneous biopsy in 96 patients or by TABLE 1: Demographics, Causes, Tumor Characteristics, and Radiofrequency Ablation (RFA) Techniques in 15 Patients With Venous Thrombosis After RFA of Hepatocellular Carcinoma Patient No. Cause of Liver Disease Liver Cirrhosis Location (Segment No.) Size (cm) Approach Type of RFA Electrode Ablation Time (min) No. of Overlapping Ablations 1 HBV Absent Percutaneous Expandable multitined HBV Present Percutaneous Internally cooled (Cluster) 27 ( ) 2 3 HBV Present Percutaneous Internally cooled (2 cm) HBV Present Intraoperative without Pringle maneuver Internally cooled (3 cm) HBV Present Percutaneous Internally cooled (3 cm) HBV Absent Percutaneous Internally cooled (3 cm) Alcoholism Present 4 and Percutaneous Internally cooled (3 cm) 24 ( ) 2 8 HBV Present Percutaneous Internally cooled (3 cm) HBV Absent Percutaneous Internally cooled (3 cm) 24 ( ) 3 10 HCV Present 5 and Percutaneous Internally cooled (Cluster and 27 ( ) 2 3 cm) 11 HBV Absent Percutaneous Expandable multitined HBV Present Intraoperative with Pringle Internally cooled (Cluster) 27 ( ) 4 maneuver 13 HCV Present Percutaneous Internally cooled (Cluster) 16 (8 + 8) 2 14 HBV Present Percutaneous Internally cooled (3 cm) HBV Present Percutaneous Internally cooled (Cluster) 12 1 Note HBV = hepatitis B virus, HCV = hepatitis C virus. AJR:197, December

3 Kim et al. imaging studies (CT or MRI) or laboratory findings (α-fetoprotein level, > 400 μg/l) in 1273 patients. Each of the patients had 1 9 nodular HCCs, with a mean of 1.39 HCCs per patient. The tumors measured cm in maximum diameter, with a mean size of 1.93 cm. All patients who were not candidates for surgical therapy received RFA if they met the following criteria: a single nodular HCC 5 cm or smaller in maximum diameter; multinodular (up to three in number) HCCs 3 cm or smaller in maximum diameter; tumors accessible via the percutaneous route at ultrasound, intraoperative, or laparoscopic approach; the absence of portal venous thrombosis and known extrahepatic metastases; Child-Pugh class A or B liver cirrhosis; prothrombin time ratio greater than 50% (prothrombin time with international normalized ratio < 1.7); and a platelet count greater than 50,000 cells/mm 3 ( cells/l). RFA Procedure and Follow-Up A total of 1379 RFAs were performed by five experienced radiologists percutaneously under realtime ultrasound guidance (n = 1289), intraoperatively (n = 85), or laparoscopically (n = 5) using internally cooled electrodes and a 200W generator (Cool-tip electrode, Covidien; Single 3 cm [n = 901], Single 2 cm [n = 301], Cluster [n = 157], or Proteus electrode [n = 19], Starmed) or using expandable multitined electrodes (LeVeen needle electrode, RadioTherapeutics) with a 460-kHz generator (RTC 2000, Boston Scientific Japan [n = 16]). In percutaneous RFA, all patients were treated while under IV conscious sedation with 50 mg of pethidine hydrochloride (Pethidine, Samsung Pharmaceutical) and local anesthesia. Patients cardiovascular and respiratory systems were continuously monitored during the procedures. Our strategy of RFA was to include the entire tumor itself and the surrounding normal hepatic parenchyma with a cm ablative margin [19, 20], regardless of RFA devices or tumor size. All patients were followed up with contrastenhanced three phase CT. The first follow-up CT scan was performed immediately and 1 month after treatment, and if the RFA achieved a technical success at this time, follow-up CT was repeated every 3 months after this visit. However, the interval of follow-up was changed if there were any suspected complications or other clinical concerns. We continued follow-up of the cases that showed venous thrombosis to determine whether there was recurrence of tumor locally beyond 4 months. Follow-up CT examinations were performed using one of five helical scanners (HiSpeed CT/I, LightSpeed QX/I, LightSpeed Ultra, and LightSpeed 16, all from GE Healthcare; and Brilliance 40, Philips Healthcare). Assessment of Venous Thrombosis We retrospectively reviewed all CT scans of the 15 patients in whom treatment-related portal or hepatic venous thrombosis developed after RFA of HCC. Two radiologists reviewed CT scans obtained before and after RFA and evaluated the frequency, morphologic patterns, sequential changes, and clinical course of portal or hepatic venous thrombosis at the RFA zone (mean follow-up, days; range, days) by consensus. Frequency We evaluated the frequency of the treatment-related portal and hepatic venous thrombosis of all the hepatic tumors treated with RFA between February 2005 and October 2009 in our institution. We also evaluated the time interval between the RFA procedure and the first CT scan on which venous thrombosis was detected. Morphologic patterns We investigated the morphologic patterns of venous thromboses, such as whether the thrombosis is located centrally or peripherally to the ablation zone, or how closely abutted it is to the zone. We also evaluated the length and level of vascular segment involved, the size of the involved vessel, and enhancement pattern of the thrombosed vessel. Sequential changes We evaluated sequential changes of venous thrombosis by assessing whether it was decreased, increased, or unchanged in 13 patients. We were unable to evaluate the sequential changes in the remaining two patients who were further treated with TACE, followed by surgery and radiation therapy, respectively, because of the recurrent tumor or intrahepatic metastasis. Clinical significance Complications were assigned to major and minor categories [21]. Major complications were defined as those that required treatment and additional hospitalization or that resulted in permanent adverse sequelae. All other complications were considered to be minor. Clinical significance was assessed by review of the patient s medical record. We evaluated the possible need for further management or hospitalization resulting from the venous thrombosis for all cases. Relationships to local tumor progression Local tumor progression was defined as the development of new tumors around the ablation zone according to the tumor enhancement patterns in three-phase liver CT, early arterial enhancement, and early washout [22, 23]. We evaluated local tumor progression in relation to the portal and hepatic venous thrombosis. Local tumor progression can be developed in three different forms: separated from the venous thrombosis, abutting the venous thrombosis, or the venous thrombosis itself (malignant thrombosis). We analyzed the local tumor progression that occurred in patients with venous thrombosis after RFA and evaluated the clinical meaning. Results The morphologic pattern, sequential change, liver function test, and presence of local tumor progression in 15 patients with venous thrombosis after RFA are summarized in Table 2. Frequency and Morphologic Pattern A total of 15 venous thromboses (1.08% of tumor occurrences), including 12 portal veins and three hepatic veins, were observed at follow-up CT after RFA. The time interval between the RFA procedure and the development of venous thromboses was days (mean, 37 days). Thromboses were found in the portal or hepatic vein branches central (n = 10), peripheral (n = 4), and both central and peripheral (n = 1) to RFA zones. Eleven cases of venous thrombosis developed in contact with the margins of RFA zones, whereas the other four cases were in close proximity to the ablation zone, measuring 2, 8, 11, and 20 mm, respectively, from the zone. The lengths of venous thromboses ranged from 10 to 74 mm, with a mean length of 35.2 mm. The minimal diameter of the thrombosed vessel ranged from 2.0 to 6.1 mm (mean, 3.65 mm). There was no linear correlation between the lengths or sizes of venous thrombosis and tumor thrombosis or local blend thrombosis. The length of the thrombosed vascular segments varied from segment 1 to segment 4, with or without dilatation of the thrombosed vessel. Sequential Changes Of 15 venous thrombosis cases, 13 were evaluated by their sequential changes. Among the 13 cases evaluated, thromboses were decreased in nine (69.2%), persisted in one (7.6%), and increased in three (23.0%). The sequential changes of the remaining two cases were not measurable because they were further treated with TACE followed by surgery and radiation therapy, respectively, because of recurrent tumor or intrahepatic metastasis. Clinical Significance Of all 15 patients who developed venous thrombosis after RFA, no patient required further treatment, such as anticoagulation or additional hospitalization, as a result of thrombosis (Figs. 1 and 2). We can therefore consider venous thrombosis after RFA as a minor complication rather than a major complication. Liver function tests performed at the time of thrombosis development were within the normal range in nine patients. The remaining six patients presented with elevat AJR:197, December 2011

4 Venous Thrombosis After Radiofrequency Ablation TABLE 2: Morphologic Pattern, Sequential Change, Liver Function Test, and Presence of Local Tumor Progression in 15 Patients With Venous Thrombosis After Radiofrequency Ablation (RFA) Minimal Diameter of Thrombosed Vessel (mm) Local Tumor Progression Segmental Dilatation Sequential Change Thrombosed Vascular Segment Distance From Ablation Margin (mm) Length (mm) Time Interval (d) Location Thrombosed Vein Patient No. 1 Portal 0 Central Decreased No Portal 28 Peripheral 11 43, 21 2, 3 2, 3 Nonassessable No Portal 0 Central , 3 2, 3 Persisted No Portal 28 Peripheral Absent Decreased Yes a Portal 30 Central , 1 2 Decreased No Portal 22 Central Decreased No Portal 22 Peripheral Absent Decreased No Portal 90 Central , 3 4, 3 Nonassessable Yes b Hepatic 20 Central and peripheral , 2 1, 2 Decreased No Hepatic 23 Central , 2, 1 Absent Decreased No Portal 128 Central , 2 1, 2 Increased Yes c Portal 6 Central Decreased No Hepatic 0 Central , 2 3, 2 Increased Yes a Portal 35 Central Absent Increased No Portal 123 Peripheral , 2 3 Decreased No 3.0 a Time interval between RFA and presence of local tumor progression was 92 days. Further treatment was transarterial chemoembolization. b Time interval between RFA and presence of local tumor progression was 90 days. Further treatment was segmentectomy. c Time interval between RFA and presence of local tumor progression was 128 days. Further treatment was palliative care. ed liver enzyme levels but showed normalization in follow-up laboratory tests within 1 month, suggesting no clinical significance. After RFA, α-fetoprotein levels decreased in six patients, levels were stable in five patients who had initial α-fetoprotein levels less than 10 μg/l, and showed a marked increase in four patients who developed tumor progression after RFA. There were no significant changes in α-fetoprotein levels at the time of venous thrombosis development. Local tumor progression was observed in four patients (26.6%) with venous thrombosis after RFA. Two patients exhibited development of abutting local tumor progression, and one patient exhibited separating local tumor progression from the venous thrombosis. The remaining patient exhibited local tumor progression developing from the venous thrombosis itself, which is considered malignant thrombosis. Among these cases, two patients were treated with TACE, and one patient was treated with segmentectomy. The remaining patient refused treatment, so palliative care was performed, resulting in tumor progression. Discussion The frequency of treatment-related venous thrombosis after RFA in the current study was 1.08% (15/1379), which is lower than the frequency found in a study by de Baère et al. [24], who reported that segmental portal vein or portal trunk thrombosis occurred in 1.7% (6/350) and hepatic vein thrombosis occurred in 1.4% (5/350) of RFA sessions. Portal or hepatic vein thrombosis after RFA is caused mainly by RFA heat damage to the endothelial cells of the portal or hepatic vein near the ablation zone [25, 26]. The vascular injury involves the intimal layer of the portal or hepatic vein, which in turn leads to platelet aggregation and the subsequent formation of thrombosis [15, 26 28]. This venous thrombosis could be the result of immediate thermal injury adjacent to the RFA site, which propagates along the distal section of the portal vein [25]. In this study, we defined ablation-related venous thromboses as those within 2 cm from the ablation margin resulting from thermal damage, which is presumed to affect adjacent vessels. However, according to a previous experimental study [27], thrombosis and coagulation of vessels smaller than 3 mm in diameter occurs commonly after RFA. Indeed, whenever RFA is performed, numerous vessels are coagulated, but it is not possible to detect this phenomenon via image-monitoring techniques in most patients. On the other hand, thrombosis of vessels larger than 4 mm after RFA is infrequent, provided that normal flow is maintained through these vessels [15, 27]. In the current study, the mean diameter of the venous thrombosis was 3.65 mm (range, mm), thus supporting the results of the prior studies. To the best of our knowledge, there have been no studies that describe the morphologic patterns or sequential AJR:197, December

5 Kim et al. changes of venous thrombosis after RFA. In this study, we found that thrombosis may develop either in central or peripheral branches of the portal or hepatic venous system, and most cases (73%) are in close contact to the margin of the RFA zone. In the thrombosed vessels, the length, thrombosed vascular segment, and presence of dilatation are variable in each case. No patients had bile duct dilatation adjacent to the thrombosed vessels. A majority of the thromboses (69.2%) displayed decreases on follow-up images. The current study also shows that portal or hepatic venous thrombosis is a minor complication after RFA and that this condition does not require hospital admission or prolonged hospital stays. There were no patients who complained of venous thrombosis related clinical symptoms or who required further management, such as anticoagulation therapy or hospitalization. de Baère et al. [24] reported 11 A A C B B D Fig year-old man with hepatocellular carcinoma in segment 4 that was treated by percutaneous radiofrequency ablation (RFA). A, Transverse helical portal venous phase CT scan obtained 1 month after RFA. Nonenhancing filling defect (arrow) is present in left main portal vein, abutting RFA zone (arrowheads) in segment 4. B, Transverse helical portal venous phase CT scan obtained 4 months after RFA. Portal venous thrombus has been disappeared. There was no need for further management or additional hospitalization in this patient. Fig year-old man with hepatocellular carcinoma in segment 8 that was treated by percutaneous radiofrequency ablation (RFA). A, Transverse helical hepatic venous phase CT scan obtained 1 month after RFA. Nonenhancing filling defect (arrow) is present in middle hepatic vein, abutting RFA zone (arrowheads) in segment 8. B, Coronal reformatted image of hepatic venous phase CT scan obtained 1 month after RFA. Nonenhancing filling defect (arrow) is present in middle hepatic vein, abutting RFA zone in segment 8 (arrowhead). Note wedge-shaped subsegmental lowattenuated area peripheral to occluded hepatic vein. C, Transverse helical hepatic venous phase CT scan obtained 4 months after RFA. Arrow indicates restoring hepatic venous flow. D, Coronal reformatted image of hepatic venous phase CT scan obtained 4 months after RFA. Hepatic venous thrombus has disappeared without need for management or additional hospitalization during follow-up period. Arrow indicates restoring hepatic venous flow. vascular thromboses after RFA for hepatic tumors, and among these, eight thrombotic events did not induce clinical symptoms or unusual modifications on blood tests and resolved spontaneously. In a separate study, a patient with portal venous thrombosis after RFA had no clinical symptoms. Additionally, this thrombosis disappeared spontaneously after 3 months without medical treatment [26]. Although most thromboses in larger vessels are asymptomatic, extensive thrombosis can result in potentially fatal complications, such as portal hypertension or hepatic failure, especially in patients with poor hepatic reserve [15, 24, 26, 29]. de Baère et al. [24] described three complete and extensive portal vein thromboses after RFA for hepatic tumors in a patient who presented with abdominal pain, ascites, and a major increase in bilirubin levels and who eventually died 4 months after treatment. Therefore, early treatment is likely to resolve the thrombus and restore normal flow in the portal vein [26], especially in patients with clinical symptoms or laboratory deterioration. When RFA is performed intraoperatively, the Pringle maneuver (hepatic inflow occlusion) has been shown to increase the size of thermal lesions produced by the procedure [30, 31]. In the present study, the Pringle maneuver was done for patient 12 during intraoperative RFA. Because hepatic blood flow generates a significant heat sink effect counteracting the RFA treatment, hepatic inflow occlusion has been shown to enhance the ablative process in animal studies [30, 32]. However, this maneuver may be risky when applied to liver tumors close to the main branch of the portal vein resulting from loss of a protective effect of hepatic blood flow on the vein wall against the thermal injury of RFA. Hence, cirrhotic livers 1478 AJR:197, December 2011

6 Venous Thrombosis After Radiofrequency Ablation might be more likely to have venous thromboses than noncirrhotic livers after RFA because of the relatively slow portal flow in cirrhotic livers. One study [24] reported that the rate of portal vein thromboses after Pringle maneuver RFA performed in cirrhotic patients (2/5) was significantly higher than the rate in noncirrhotic livers (0/54). Therefore, we hypothesize that relatively low blood flow or occlusion of blood flow can precipitate the occurrence of venous thrombosis after RFA. In addition, recent studies have suggested that the Pringle maneuver should be avoided when RFA is applied close to the main portal vein branch [25]. We identified four patients with local tumor progression among patients who presented with portal or hepatic venous thrombosis after RFA. We also observed malignant tumor thrombus in one patient after RFA. Of the treated patients, the local tumor progression rate around the RFA site ranged widely, from 3.6% to 20%. However, a slightly higher rate of local tumor progression (26.6% [4/15]) was seen in patients with venous thrombosis after RFA in this study, compared with the rate of local tumor progression (3.6 20%) in patients irrespective of venous thrombosis after RFA in prior studies [33, 34]. At present, it is unclear whether such local tumor progression is directly related to the RFA procedure or to the occurrence of venous thrombosis. However, it is implied that RFA therapy may increase the risk of venous invasion of HCC tumors. Takada et al. [29] reported two cases of recurrence with massive portal tumor thrombus around the treated area after RFA for HCCs. We speculate that an increase in the intratumoral pressure, resulting from elevated temperatures during the ablation process, may induce the release of tumor cells. Another possibility is that the penetration of deployed expandable needles into a small portal branch may lead to intraportal tumor shedding [17, 29, 35]. Our study has several limitations. First, the retrospective study design might cause selection bias. Second, there is heterogeneity in the treatment procedures, including percutaneous, intraoperative, and laparoscopic RFA, using various kinds of RFA electrodes. Third, there was lack of pathologic correlation between malignant and benign thrombosis. Finally, because of the small sample size, we did not analyze local tumor progression in patients without venous thrombosis after RFA. In conclusion, the development of portal or hepatic venous thrombosis after RFA in patients with HCC is rare and, because it usually results in favorable prognoses, is considered a minor complication in most cases. Further investigations are warranted to elucidate whether venous thromboses after RFA are related to local tumor progression around RFA zones or within the venous system. References 1. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology 2005; 42: Cho YK, Kim JK, Kim MY, Rhim H, Han JK. Systematic review of randomized trials for hepatocellular carcinoma treated with percutaneous ablation therapies. Hepatology 2009; 49: Brunello F, Veltri A, Carucci P, et al. Radiofrequency ablation versus ethanol injection for early hepatocellular carcinoma: a randomized controlled trial. Scand J Gastroenterol 2008; 43: Rossi S, Garbagnati F, Rosa L, Azzaretti A, Belloni G, Quaretti P. Radiofrequency thermal ablation for treatment of hepatocellular carcinoma. Int J Clin Oncol 2002; 7: Choi D, Lim HK, Rhim H, et al. Percutaneous radiofrequency ablation for early-stage hepatocellular carcinoma as a first-line treatment: long-term results and prognostic factors in a large single-institution series. Eur Radiol 2007; 17: Curley SA, Izzo F, Ellis LM, Nicolas Vauthey J, Vallone P. Radiofrequency ablation of hepatocellular cancer in 110 patients with cirrhosis. Ann Surg 2000; 232: Livraghi T. Percutaneous ethanol injection in the treatment of hepatocellular carcinoma in cirrhosis. Hepatogastroenterology 2001; 48: Shibata T, Iimuro Y, Yamamoto Y, et al. Small hepatocellular carcinoma: comparison of radio-frequency ablation and percutaneous microwave coagulation therapy. Radiology 2002; 223: Izzo F. Other thermal ablation techniques: microwave and interstitial laser ablation of liver tumors. Ann Surg Oncol 2003; 10: Adam R, Akpinar E, Johann M, Kunstlinger F, Majno P, Bismuth H. Place of cryosurgery in the treatment of malignant liver tumors. Ann Surg 1997; 225:39 38, discussion Llovet JM, Bruix J. Systematic review of randomized trials for unresectable hepatocellular carcinoma: chemoembolization improves survival. Hepatology 2003; 37: Bruix J, Sala M, Llovet JM. Chemoembolization for hepatocellular carcinoma. Gastroenterology 2004; 127:S179 S Rossi S, Di Stasi M, Buscarini E, et al. Percutaneous radiofrequency interstitial thermal ablation in the treatment of small hepatocellular carcinoma. Cancer J Sci Am 1995; 1: Livraghi T, Goldberg SN, Lazzaroni S, Meloni F, Solbiati L, Gazelle GS. Small hepatocellular carcinoma: treatment with radio-frequency ablation versus ethanol injection. Radiology 1999; 210: Rhim H. Complications of radiofrequency ablation in hepatocellular carcinoma. Abdom Imaging 2005; 30: Rhim H, Yoon KH, Lee JM, et al. Major complications after radio-frequency thermal ablation of hepatic tumors: spectrum of imaging findings. Radio- Graphics 2003; 23: , discussion Kawamoto C, Yamauchi A, Baba Y, Kaneko K, Yakabi K. Measurement of intrahepatic pressure during radiofrequency ablation in porcine liver. J Gastroenterol 2010; 45: Farmer DG, Rosove MH, Shaked A, Busuttil RW. Current treatment modalities for hepatocellular carcinoma. Ann Surg 1994; 219: Dodd GD 3rd, Frank MS, Aribandi M, Chopra S, Chintapalli KN. Radiofrequency thermal ablation: computer analysis of the size of the thermal injury created by overlapping ablations. AJR 2001; 177: Goldberg SN, Charboneau JW, Dodd GD 3rd, et al. Image-guided tumor ablation: proposal for standardization of terms and reporting criteria. Radiology 2003; 228: Goldberg SN, Grassi CJ, Cardella JF, et al. Imageguided tumor ablation: standardization of terminology and reporting criteria. Radiology 2005; 235: Catalano O, Lobianco R, Esposito M, Siani A. Hepatocellular carcinoma recurrence after percutaneous ablation therapy: helical CT patterns. Abdom Imaging 2001; 26: Nakazawa T, Kokubu S, Shibuya A, et al. Radiofrequency ablation of hepatocellular carcinoma: correlation between local tumor progression after ablation and ablative margin. AJR 2007; 188: de Baère T, Risse O, Kuoch V, et al. Adverse events during radiofrequency treatment of 582 hepatic tumors. AJR 2003; 181: Ng KK, Lam CM, Poon RT, Shek TW, Fan ST, Wong J. Delayed portal vein thrombosis after experimental radiofrequency ablation near the main portal vein. Br J Surg 2004; 91: Zheng RQ, Kudo M, Inui K, et al. Transient portal vein thrombosis caused by radiofrequency ablation for hepatocellular carcinoma. J Gastroenterol 2003; 38: Lu DS, Raman SS, Vodopich DJ, Wang M, Sayre J, Lassman C. Effect of vessel size on creation of hepatic radiofrequency lesions in pigs: assessment of the heat sink effect. AJR 2002; 178: Ng KK, Lam CM, Poon RT, Fan ST. Portal vein thrombosis after radiofrequency ablation for recurrent hepatocellular carcinoma. Asian J Surg 2003; 26:50 53, discussion 54 AJR:197, December

7 Kim et al. 29. Takada Y, Kurata M, Ohkohchi N. Rapid and aggressive recurrence accompanied by portal tumor thrombus after radiofrequency ablation for hepatocellular carcinoma. Int J Clin Oncol 2003; 8: Chinn SB, Lee FT Jr, Kennedy GD, et al. Effect of vascular occlusion on radiofrequency ablation of the liver: results in a porcine model. AJR 2001; 176: Rossi S, Garbagnati F, De Francesco I, et al. Relationship between the shape and size of radiofrequency induced thermal lesions and hepatic vascularization. Tumori 1999; 85: Patterson EJ, Scudamore CH, Owen DA, Nagy AG, Buczkowski AK. Radiofrequency ablation of porcine liver in vivo: effects of blood flow and treatment time on lesion size. Ann Surg 1998; 227: Llovet JM, Vilana R, Bru C, et al. Increased risk of tumor seeding after percutaneous radiofrequency ablation for single hepatocellular carcinoma. Hepatology 2001; 33: Shiina S, Teratani T, Obi S, et al. A randomized controlled trial of radiofrequency ablation with ethanol injection for small hepatocellular carcinoma. Gastroenterology 2005; 129: Seki T, Tamai T, Ikeda K, et al. Rapid progression of hepatocellular carcinoma after transcatheter arterial chemoembolization and percutaneous radiofrequency ablation in the primary tumour region. Eur J Gastroenterol Hepatol 2001; 13: FOR YOUR INFORMATION The American Roentgen Ray Society now provides instant Web exclusive access to its annual meeting abstracts. The abstracts, featured as a supplement to the AJR, summarize the latest comprehensive and clinically important information presented at ARRS s annual meetings. The abstracts can be viewed online by visiting AJR:197, December 2011