fax: (81) 3-3814-0021.
Liver Failure and Liver Disease
Radiofrequency ablation for hepatocellular carcinoma in so-called high-risk locations†
Article first published online: 20 APR 2006
Copyright © 2006 American Association for the Study of Liver Diseases
Volume 43, Issue 5, pages 1101–1108, May 2006
How to Cite
Teratani, T., Yoshida, H., Shiina, S., Obi, S., Sato, S., Tateishi, R., Mine, N., Kondo, Y., Kawabe, T. and Omata, M. (2006), Radiofrequency ablation for hepatocellular carcinoma in so-called high-risk locations. Hepatology, 43: 1101–1108. doi: 10.1002/hep.21164
Potential conflict of interest: Nothing to report.
- Issue published online: 20 APR 2006
- Article first published online: 20 APR 2006
- Manuscript Accepted: 13 FEB 2006
- Manuscript Received: 16 APR 2005
We evaluated the efficacy and safety of radiofrequency (RF) ablation for hepatocellular carcinoma (HCC) in presumably high-risk locations. Between February 1999 and December 2001, we performed RF ablation on 1,419 nodules in 636 consecutive HCC patients, of which 231 nodules in 207 patients were in high-risk locations, defined as less than 5 mm from a large vessel or an extrahepatic organ. Eighty-one patients had a nodule adjacent to a large vessel, 145 patients had a nodule adjacent to an extrahepatic organ, of whom 19 also had one adjacent to a large vessel. Early complications and local tumor progression were analyzed with regard to the location of each nodule. The mean nodule diameter and average number per patient were 27 mm and 2.3, respectively. Early complications, within 30 days after ablation, occurred in 12 of 207 patients (5.8 %) with a nodule in a high-risk location and in 15 of 429 patients (3.5 %) without (P = .1776). There was no significant difference in local tumor progression rate between nodules in high-risk locations (1 year: 2.1%, 2 years: 3.1%, 3 years: 3.1%) and those elsewhere (1 year: 0.6%, 2 years: 1.7%, 3 years: 2.5%) (P = .2745). In conclusion, HCC nodules adjacent to a large vessel or extrahepatic organ were treated with RF ablation without compromising the efficacy of the procedure. However, even though without significant difference, some complications occurred at risky locations and need to be carefully considered. (HEPATOLOGY 2006;43:1101–1108.)
Hepatocellular carcinoma (HCC) is one of the most common malignancies in the world, causing more than 420,000 deaths every year.1 Its incidence has been increasing worldwide2–5 due to the spread of hepatitis C virus (HCV) infection. The majority of patients with HCC have cirrhosis,6, 7 and impaired liver function as well as multiplicity of lesions may contraindicate curative surgical resection. Although orthotopic liver transplantation offers the best chance for therapeutic success,8 its performance is limited by a shortage of donor organs. Furthermore, HCV infection, the dominant cause of HCC in various regions including Japan, recurs after transplantation, leading to severe liver damage.9, 10 In the meantime, nonsurgical treatments such as percutaneous ethanol injection therapy (PEIT)11–13 and percutaneous microwave coagulation therapy (PMCT)14 have played an important role, based on their capability to be used for a wider range of patients.
Radiofrequency (RF) ablation was recently introduced as a novel ablation modality for HCC. Several investigators have reported that percutaneous RF ablation for patients with small HCC nodules provides favorable survival with excellent local control,15, 16 and it may also be used as a bridge to liver transplantation.17, 18 The RF current emitted from the electrode is converted into heat and necrotizes the tumor. Therapeutic efficacy is considered to be more predictable with RF ablation than with PEIT, as the latter depends on diffusion of liquid ethanol in tissue. Although RF ablation is fairly safe in general,19 a broad spectrum of complications has been reported in large-scale multicenter surveys. Some investigators have suggested that tumor location was closely related to the risk of major complications. Central tumors close to the hepatic hilum were reported to be unsuitable for percutaneous RF ablation because of the risk of injuring adjacent bile ducts.20 It was also suggested that RF ablation for nodules adjacent to large vessels may often result in incomplete necrosis because of a heat sink effect. In addition, peripheral tumors adjacent to extrahepatic organs were also suggested to be unsuitable because of the risk of heat injury, such as intestinal perforation and pleural effusion.21, 22 Thus, there may be a difficulty with RF ablation of nodules in such high-risk locations, possibly resulting in complications or not allowing sufficient treatment.
In the authors' institution, we have placed no restrictions on RF ablation for liver tumor solely by its location, and ablation even for nodules close to the hepatic hilum or adjacent to extrahepatic organs has been performed. To confirm the safety and effectiveness of the procedure, we retrospectively evaluated early complications and local tumor progression as an indicator of insufficient treatment in 636 followed-up HCC patients treated with RF ablation from the perspective of tumor location.
Patients and Methods
The indication criteria of RF ablation for HCC were as follows: (1) lesions were surgically unresectable or the patient voluntarily chose nonsurgical treatment, (2) total serum bilirubin concentration was lower than 3 mg/dL, and (3) no extrahepatic metastasis or vascular tumor invasion was observed. In general, we performed RF ablation on patients with three or fewer lesions, all of which were 3 cm or less in diameter. However, we also performed RF ablation on patients with more than three lesions or lesions larger than 3 cm in diameter if the procedure could be assumed to be clinically effective. Patients were excluded if they had excessive bleeding tendency (platelet count below 50 × 109/L or prothrombin activity below 50%), or refractory ascites. No patients were excluded solely because of the location of lesions. The study was performed according to the guidelines of the Helsinki Declaration. A written, fully informed consent was obtained from each patient before treatment.
Diagnosis of Hepatocellular Carcinoma.
Pretreatment work-up for the diagnosis of HCC included ultrasonography, dynamic helical computed tomography (CT), and CT during arterial portography and hepatic arteriography.23–25 Diagnosis of HCC was confirmed by ultrasound-guided tumor biopsy using a 20-gauge needle (Monopty, C.R.Bard, Inc., Convington, GA) in 477 cases (75%). Histological grade of tumor differentiation was determined according to the modification of the Edmondson Grading System proposed by the Liver Cancer Study Group of Japan.26, 27 In the remaining 159 cases, diagnosis was based on typical imaging findings, i.e., arterial-phase hyperattenuation and late-phase contrast washout on dynamic CT.28 Tumor location was defined according to the Couinaud nomenclature by using ultrasonography.29
On the basis of previous literature, we defined locations adjacent to large vessels or extrahepatic organs as high-risk locations. HCC nodules adjacent to large vessels were defined as those located less than 5 mm from a first or second branch of the portal vein, the base of hepatic veins, or the inferior vena cava, while nodules adjacent to extrahepatic organs were defined as those located less than 5 mm from the heart, lung, gallbladder, right kidney, or gastrointestinal tract. The distance between the edge of the nodule and the large vessel or extrahepatic organ was measured on CT images reconstructed at 5-mm intervals.
A curative case was defined as one in which all HCC nodules were intended to be treated by RF ablation, thus excluding those with a nodule left untreated even if other nodules were treated by RF ablation with the intent of reducing tumor burden. Local tumor progression was defined as the appearance of viable tumor during follow-up that was contiguous with the zone that had been considered completely ablated.30 Early complication was defined as that occurring within 30 days after RF ablation and requiring additional invasive therapy and/or lengthened hospitalization more than one week.
RF Ablation Technique and Equipment.
Percutaneous RF ablation was performed on an inpatient basis. First, two large grounding pads with a surface area of 400 cm2 were attached to the thighs. Pentazocine (30 mg), hydroxyzine (25 mg), and atropine (0.5 mg) were administered intravenously. We used a 480-kHz monopolar RF generator (CC-1 Cosman Coagulator, Radionics, Burlington, MA) capable of producing up to 150 W power, and a 20-cm long, 18-gauge, internally-cooled-tip RF electrode with a 2- or 3-cm long exposed metallic tip (Radionics). After local anesthesia, the electrode was inserted under ultrasound guidance (Power Vision 8000, Toshiba, Tokyo, Japan). During ablation, a thermocouple embedded in the electrode tip continuously monitored the local temperature. Tissue impedance was also monitored continuously by circuitry incorporated in the generator. A peristaltic pump (Watson-Marlow, Wilmingtom, MA) was used to infuse 0°C normal saline into the cooling lumen of the electrode at a rate sufficient to maintain tip temperature below 20°C. In case of 3-cm tip electrodes, the output started at 60 W, and was increased at 20 W/min until tissue impedance overshot. Then, the output was decreased by 20 W and maintained for 12 min. Impedance overshooting sometimes occurred two times or more in a session. With 2-cm tip electrodes, the output started at 40 W, was increased at 20 W/minutes until impedance overshooting, and then maintained for 6 minutes. The ablation zone was discernible as a transient hyperechoic zone under ultrasound during the procedure. Correlation between hyperechoic zone and necrotic area has been pathologically confirmed.31 A 12-minute ablation using a 3-cm tip electrode was assumed to necrotize a sphere of 3 cm in diameter. For large nodules, the electrode was repeatedly inserted into different sites (overlapping ablations), so that the entire nodule was enveloped in assumed necrotic volumes.
Ablation in High-Risk Locations.
When a nodule was in a high-risk location as defined above, we first carefully considered the route of electrode insertion on ultrasound scrutiny. The route was selected so as not to injure the lung, gastrointestinal tract, gall bladder, portal vein, IVC, and bile duct. When the targeted nodule was close to the diaphragm, we used an artificial pleural effusion method with 5% glucose to separate the lung,32 when it was close to the gastrointestinal tract, we infused 5% glucose (250-3000 mL) as an artificial ascites into the abdominal cavity to separate the gastrointestinal tract to prevent thermal injury (Fig. 1). Second, we adjusted the ablation time, stopping ablation the instant microbubbles generated around the electrode tip touching the adjacent large vessel or extrahepatic organ (Figs. 1 and 2). All RF ablation procedures were performed by one of seven experienced physicians (S.S., T.T., K.H., T.F., S.S., Y.K., M.A.) with at least 4 years of experience in performing ultrasound-guided percutaneous ablation, including PEIT and PMCT.
Assessment of Treatment and Follow-Up.
Treatment response was assessed by contrast-enhanced spiral CT at 1 to 3 days after the end of the treatment. Complete response was confirmed by the absence of enhanced areas. After the treatment, tumor biomarkers, serum alpha-fetoprotein, lectin-reactive alpha-fetoprotein fraction, and des-gamma-carboxy-prothrombin were measured every month, and helical CT and ultrasonography were performed every 3 to 4 months. When recurrence was suspected, imaging studies and biopsy were performed.
Difference in means was analyzed by Student's t-test and frequency distribution was compared by chi-square test. Cumulative incidence of local tumor progression was calculated using the Kaplan-Meier technique, and differences were tested using the log-rank test. Variables associated with local tumor progression were assessed by multivariate Cox proportional hazard model, including tumor location, tumor size, number of procedures, and duration of ablation. A P value of less than .05 by two-tailed analysis was considered to be significant. Data processing and analysis were performed using the SAS system (SAS Institute Inc., Cary, NC).
Patients and HCC Nodules.
Between February 1999 and December 2001, 795 patients with HCC were treated at the Department of Gastroenterology, University of Tokyo Hospital. Among them, 636 were treated with percutaneous RF ablation (clinical profiles shown in Table 1), 224 for primary HCC and the remaining 412 for recurrent HCC. The number of nodules exceeded three in 120 patients (19%), and the nodule diameter was larger than 50 mm in 37 patients (6%). Curative RF ablation was intended in 597 (93.9%) of 636 patients, and a total of 1243 nodules were ablated. In the remaining 39 patients, RF ablation was intended for a total of 176 nodules to reduce tumor burden, with some nodules being left unablated in each patient because of their multiplicity. Thus, a total of 1,419 nodules in 636 patients were treated with ablation. Among them, 231 nodules were in a high-risk location as defined above (Table 3), and a total of 207 patients (32.5%) had at least one nodule in a high-risk location.
|Age (years)||66.9 ± 0.3|
|HBs Ag (+)||76 (12%)|
|HCV Ab (+)||524 (82%)|
|Biochemistry (mean ± SD)|
|ALT (IU/L)||58 ± 35|
|T.Bil (mg/dl)||1.0 ± 0.5|
|Alb (g/dl)||3.5 ± 0.5|
|PT (%)||77 ± 13|
|Indication of treatment|
|Primary HCC||224 (35%)|
|Recurrence HCC||412 (65%)|
|AFP < 20 (ng/mL)||285 (45%)|
|Tumor size (mm: mean ± SD)||27 ± 12|
|− 20 (mm)||221 (35%)|
|> 51||37 (6%)|
|Number of nodules||2.3 ± 1.7|
|Treatment prior to RFA|
|Hepatic resection||49 (8%)|
|Nodule Adjacent to Large Vessel (n = 79)||Nodule Adjacent to Extrahepatic Organ (n = 62)|
|Base of the|
Outcome of RF Ablation.
We treated 1417 (99.9%) of the 1419 nodules completely with RF ablation, as confirmed by follow-up CT. In one case, we gave up ablation for a nodule abutting on the heart that was difficult to identify by ultrasound. In the other case, we could not find a safe route of electrode insertion for a nodule located in the caudate lobe of a severely atrophied liver. Thus, 229 (99.1%) of the 231 nodules in high-risk locations and all of the other 1,188 nodules were successfully treated. As shown in Table 2, nodules in high-risk locations required a larger number of treatment sessions and longer duration of ablation than those in other locations. The nodules in high-risk locations were significantly larger than the others.
|(No. of Cases) Nodules Adjacent to —||Nodules in High-Risk Locations (231)||Others (1188)||P|
|Large Vessel (69)||Extrahepatic Organ (152)||Both Large Vessel and Extrahepatic Organ (10)|
|Size (mm)||27 ± 15||27 ± 13||25 ± 14||53 ± 20||19 ± 9||< .001*|
|Number of procedures||4.0||4.2||3.6||7.9||2.1||< .001*|
|Total duration (min)||34||34||31||71||16||< .001*|
The incidence rate of early complications per patient was calculated (Table 4). There were 12 cases (5.8%) of early complications among 207 patients with at least one nodule in high-risk locations, while there were 15 (3.5%) among 429 patients without, the difference was not significant (P = .1776). Massive pleural effusion and gastrointestinal perforation occurred only in patients with nodules in high-risk locations (Fig. 3). Hemorrhage was encountered in three patients without high-risk nodules and one patient with a nodule adjacent to extrahepatic organ but not to large vessels. Emergent laparotomy was necessary in one case of bile peritonitis. In that case, RF ablation was performed in spite of the presence of intrahepatic bile duct dilation. In the remaining cases, patients recovered without surgical measures. No deaths were directly associated with complications of RF ablation.
|(No. of cases) Cases With Nodules Adjacent to —||Cases With Nodules in High-Risk Locations (207)||Others (429)||P|
|Large Vessel (62)||Extrahepatic Organ (126)||Both Large Vessel and Extrahepatic Organ (19)|
|Liver abscess penetrating into GI||1||1||0|
|Total||12 (5.8%)||3 (4.8%)||8 (6.3%)||2 (11%)||15 (3.5%)||.1776*|
Bile Ducts Injury.
Intrahepatic bile duct dilation appeared with a mean interval of 7.3 months (0.3-26 months) after RF ablation, which was presumably caused by biliary stricture in association with ablation. Although the detection of dilation was often delayed, the injury itself may have been directly caused by ablation and we included this complication in this study. Bile duct dilation occurred in 28 (2.0%) of the overall patients. There was significant difference in the incidence of bile ducts injury: 6 (7.6%) of 79 nodules near a large portal vein and 22 (1.6%) of the remaining 1340 nodules (P = .0002 by chi-square test).
Local Tumor Progression.
Local tumor progression was evaluated only among the curative cases. Among the 597 patients in whom all nodules were to be ablated, twelve were lost to follow-up before the first routinely performed follow-up CT 3 to 4 months after RF ablation. Thus, local tumor progression was investigated in 1,243 nodules of 585 patients, and 20 (1.6%) nodules were positive. There was no significant difference in the cumulative incidence of local tumor progression among the nodules in high-risk locations: 1 year: 2.1%, 2 years: 3.1%, 3 years: 3.1%; among those nodules found elsewhere: 1 year: 0.6%, 2 years: 1.7%, 3 years: 2.5% (P = .2745 by log-rank test, Fig. 4). The nodules adjacent to both large vessels and those adjacent to extrahepatic organs showed similar local tumor progression rates (P = .6418 by log-rank test, Fig. 5). We assessed four variables possibly associated with local tumor progression by multivariate Cox proportional hazard model, namely, tumor location (high-risk location vs. elsewhere), tumor size (≥3 cm vs. <3 cm), number of treatment sessions (≥5 vs. <5), and duration of ablation (≥25 minutes vs. <25 minutes). Only tumor size was significantly associated with the risk of local tumor progression, whereas tumor location was not a significant risk factor (Table 5).
|Variables||P||Hazard ratio (95% CI)|
|Tumor location (high risk)||0.8499||0.905 (0.321–2.552)|
|Tumor size (≥3 cm)||<0.0001||12.457 (3.819–40.636)|
|Number of sessions (≥5)||0.5526||1.464 (0.416–5.155)|
|Total duration (≥25 min.)||0.8581||0.884 (0.230–3.407)|
In this study, we found no significant difference in the early complication rate nor in the local tumor progression rate between RF ablation of HCC nodules in high-risk locations and that of nodules elsewhere. Complete ablation was achieved in 229 of 231 nodules in high-risk locations and all of the 1,188 nodules found elsewhere. The treatment success rate, higher than those reported by others,16, 34, 35 was probably the result of our accumulated experience in percutaneous ablation procedures with ethanol, microwave, and RF, amounting to a total of 4,000 cases.33 We have perfected techniques for inserting an electrode into tumors almost anywhere in the liver without puncturing large vessels or bile ducts by, for example, changing the patient's position and selecting the insertion site. In addition, we place no restrictions on the number of treatment sessions. We have repeated RF ablation until necrosis of the entire tumor was confirmed on CT. Although this may have increased the number of treatment sessions, complete ablation is possibly associated with improved prognosis.36
While there was no difference in early complication rates according to tumor location, the overall early complication rate of 4.2% in this study may be slightly higher than those reported in other studies.21, 37 The effort of thorough ablation increased the total number of electrode insertions, and this may have led to an increase in complications. Nevertheless, we believe that this will be well compensated by the improved prognosis resulting from complete ablation. Local tumor progression rates were not significantly different according to tumor location, indicating that the ablation of nodules in presumably high-risk locations was performed without compromising quality. Moreover, this demonstrated that the heat sink effect of large vessels on the ablation of neighboring nodules can be managed by careful ablation. According to our experience, careful real-time monitoring of microbubbles generated around the electrode tip was one of the most helpful techniques. We have made it a rule to stop ablation the instant microbubbles reached adjacent large vessels or extrahepatic organs.
We used the cool-tip electrode throughout this study. Another type of RF ablation electrode with expandable hooks is also widely used.38 Although we have not compared the two types, we are afraid that the extension of the hooks might not be precisely controllable when applied to nodules adjacent to large vessels or extrahepatic organs. Curley et al. performed RF ablation with an expandable-type electrode, and adopted a laparotomic or laparoscopic approach when ablating nodules near the liver capsule to avoid thermal injury to adjacent structures such as the diaphragm and bowels.15 As shown above, the percutaneous approach is applicable to nodules in any location in the liver when using the cool-tip electrode, although we did sometimes use the artificial ascites technique to avoid thermal injury to the bowels.
Heat injury to adjacent bile ducts remains a problem because, in contrast to blood flow, bile flow is slow and has little cooling effect. Damage to bile ducts may appear 3 or more months after RF ablation, usually in the form of dilation in the upstream bile ducts. We showed in this study that the incidence of bile duct dilation differ significantly according to the distance between the targeted tumor and intrahepatic bile duct, as surrogated by neighboring portal vein. The influence of bile duct injury on long-term prognosis remains to be studied. Tumor seeding is another major late complication of RF ablation. Although we did not analyze this in the present study, it is not likely to be related to the adjacency to large vessels or extrahepatic organs.
It should be mentioned that the outcome of the RF ablation procedure is heavily dependent on the expertise of the operators. A less-skilled operation may cause either incomplete ablation or a higher incidence of complications, resulting in inferior prognosis. As in the case of surgery, any institution engaged in RF ablation should publicize the survival rate of its patients.
In conclusion, it can be stated with confidence that RF ablation can be performed effectively on nodules adjacent to large vessels or extrahepatic organs, with the proviso that the operators possess sufficient experience and skill.
- 20Radiofrequency ablation of the liver: current status. Am J Gastroenterol 2001; 176: 3–16., .
- 27The Liver Cancer Study Group of Japan. The General Rules for the Clinical and Pathological Study of Primary Liver Cancer (in Japanese). 3rd editon. Tokyo, Japan: Kanehara, 1992.
- 32Percutaneous sonographically guided radiofrequency ablation with artificial pleural effusion for hepatocellular carcinoma located under the diaphragm. Am J Roentgenol 2004; 183: 538–588., , , , , , et al.