Radiofrequency Ablation of Hepatic Metastases - Springer Link

Ann Surg Oncol (2014) 21:3090–3095 DOI 10.1245/s10434-014-3738-y

ORIGINAL ARTICLE – HEPATOBILIARY TUMORS

Radiofrequency Ablation of Hepatic Metastases: Factors Influencing Local Tumor Progression Chang-Hsien Liu, MD1, Chih-Yung Yu, MD1, Wei-Chou Chang, MD1, Ming-Shen Dai, MD, PhD2, Cheng-Wen Hsiao, MD3, and Yu-Ching Chou, PhD4 1

Department of Radiology, Tri-Service General Hospital and National Defense Medical Center, Taipei, Taiwan, R.O.C.; Division of Oncology, Department of Internal Medicine, Tri-Service General Hospital and National Defense Medical Center, Taipei, Taiwan, R.O.C.; 3Division of Colorectal Surgery, Department of Surgery, Tri-Service General Hospital and National Defense Medical Center, Taipei, Taiwan, R.O.C.; 4School of Public Health, National Defense Medical Center, Taipei, Taiwan, R.O.C. 2

ABSTRACT Background. Although radiofrequency ablation (RFA) of nonresectable hepatic metastases has gained wide acceptance by showing survival benefit in selected patients, scattered reports are available regarding risk factors of local control of percutaneous RFA. The purpose of this study was to prospectively evaluate the factors influencing local tumor progression after percutaneous RFA of hepatic metastases. Methods. Sixty-nine hepatic metastatic lesions in 54 patients were treated by percutaneous RFA. Efficacy was evaluated by contrast-enhanced computed tomography or magnetic resonance imaging at 1 month after ablation, then at 3-month intervals for the first year and biannually thereafter. Results. The results of the log-rank test showed that tumor size of \3 cm (p = 0.024) and the absence of tumor contiguous with large vessels (p = 0.002) significantly correlated with local control for hepatic metastases. Cox regression analysis showed that the tumor size \3 cm and the absence of tumor contiguous with large vessels were independent factors (p = 0.055 and 0.009, respectively). The results of the log-rank test showed that neither the threshold post-ablation margin of 1.8 cm (p = 0.064) nor the presence of a tumor with subcapsular location (p = 0.134) correlated with the success of local control.

Conclusions. Percutaneous RFA is more effective in achieving local control in patients with hepatic metastases when the tumor size is\3 cm and not contiguous with large vessels.

Hepatic metastases, primarily from colorectal cancer, are the most frequent liver tumors.1,2 They are 18–40 times more common than primary liver neoplasm.3 Surgical resection with curative intent is the standard treatment when metastatic disease is limited to the liver; however, only 5 to 20 % of patients can benefit from resection because of medical comorbidities and/or insufficient liver function.4 Percutaneous radiofrequency ablation (RFA) is being widely used for the treatment of primary or metastatic hepatic malignancies, especially from colorectal cancer, since 1990.1,2,4–8 RFA has shown potential as a promising treatment for the local control of tumors in cases with minimal invasiveness and low morbidity and mortality. Nonetheless, incomplete treatment and strong local tumor progression remain as clinical challenges.2 Several risk factors for local tumor progression have been reported in recent studies.5–8 To our knowledge, limited data regarding potential risk factors of percutaneous RFA of hepatic metastases have been published so far. Thus, we undertook this study to prospectively evaluate the risk factors associated with local tumor progression after percutaneous RFA of hepatic metastases. MATERIALS AND METHODS

Ó Society of Surgical Oncology 2014 First Received: 28 October 2013; Published Online: 1 May 2014 C.-H. Liu, MD e-mail: [email protected]

Patient Demographics and Medical Record Review The institutional review board for studies on human subjects granted approval to conduct this study, with informed consent.

Radiofrequency Ablation of Hepatic Metastases

From January 2009 to August 2012, a total of 70 patients with inoperable hepatic metastases, determined by the number or bilobar location of the tumors, the presence of inadequate residual liver volume if resected, or inability to obtain a marginnegative resection were selected to undergo image-guided percutaneous RFA at a single institution. Sixteen patients were excluded either because of the lack of imaging follow-up after the percutaneous RFA procedure (n = 10) or because some of these patients had residual tumors and did not undergo repeat percutaneous RFA treatment at 1-month follow-up (n = 6). The remaining 54 patients with 69 metastatic nodules had the following origin of hepatic metastases: 20 colorectal carcinomas, 10 breast carcinomas, 4 pancreatic carcinomas, 4 cholangiocarcinomas, 3 lung carcinomas, 3 gallbladder carcinomas, 2 ovarian carcinomas, 2 esophageal carcinomas, 2 renal cell carcinomas, 2 leiomyosarcomas, 1 gastric carcinoma, and 1 melanoma. The criteria for percutaneous RFA eligibility were as follows: up to 5 nodules with diameters up to 8 cm, absence of extrahepatic disease, platelet count C50,000/mm3, and prothrombin activity of C50 %. Ablation Procedure Percutaneous RFA was performed by one of the two interventional radiologists with 8 and 12 years of experience in percutaneous RFA of hepatic tumors. Two RFA systems were used. A 200 W generator with straight internally cooled single electrode was used in 35 patients with 45 lesions (Covidien Inc., Burlington, MA, USA). A 200 W generator with multitined, 3-cm expandable electrodes (LeVeen electrode) was used in 19 patients with 24 lesions (Boston Scientific, Waltham, MA, USA). The choice of the system was at the discretion of the operator. Percutaneous RFA was performed with either computed tomography (CT; n = 40) or ultrasound (n = 14) guidance. All RFA procedures were performed with either intravenous conscious sedation (n = 20) or general anesthesia (n = 34). For tumors \2 cm in diameter, a single ablation was performed. For tumors C2 cm, multiple overlapping ablations were performed. The RFA algorithm for the internally cooled electrodes consisted of a pulse current, with each ablation lasting 12 min. With the 3-cm expandable electrodes, an impedance-based algorithm was used, and the average time for each ablation was 7–12 min.

3091

tumor nodule in axial, coronal, and sagittal planes was considered as a criterion for complete treatment. Thereafter, follow-up imaging was performed every 3 months for 1 year followed by semiannual follow-up imaging. The efficacy of treatment was defined as the cumulative rate of local tumor progression at follow-up imaging. Image Analysis The results of each CT and MRI study were evaluated retrospectively by a consensus reading by two abdominal radiologists with 7 and 10 years of experience in abdominal imaging. The data analyzed included tumor diameter, whether the tumor was contiguous with large vessels (a portion of the tumor located within 5 mm of an intrahepatic vessels with a diameter of[3 mm), a subcapsular location (a portion of the tumor located within 5 mm of the liver capsule), a diameter of the zone of ablation, and a post-ablation margin. The evolution of the diameter of the zone of ablation was based on the 1month follow-up image study and was reported as the largest diameter of the non-enhancing zone in an axial, coronal, or sagittal plane where the zone of ablation was the largest. The postablation margin was defined qualitatively as the thickness of a previously normal liver parenchyma that in post-ablation imaging was included in the zone of ablation. Quantitatively, the postablation margin was calculated as the difference between the maximum diameter of the post-ablation zone and the maximum pre-treatment tumor diameter, divided by 2. Nodular or irregular contrast enhancement within or at the margin of the zone of ablation during portal venous phase imaging after the 1-month follow-up examination was considered as criteria of local tumor progression. Statistical Analysis Continuous data were expressed as mean ± SD. The Kaplan–Meier method was used to estimate the interval from RFA treatment to local tumor progression. The variables that were determinants of local tumor progression were analyzed by the log-rank test. A Cox proportional hazard regression model was used to analyze independent risk factors. All p values were from two-tailed tests. A p value \0.05 indicated a statistically significant difference. The analyses were performed by SPSS software for Windows (IBM, Armonk, NY, USA).

Treatment Course and Follow-up All patients underwent immediate follow-up CT to evaluate immediate complication after percutaneous RFA. Followup contrast-enhanced CT or magnetic resonance imaging (MRI) of the abdomen was performed 1 month after the ablation to evaluate the zone of ablation. A non-enhancing zone of ablation with a diameter greater than that of the hepatic

RESULTS Effect of Treatment on Hepatic Metastases Table 1 shows the patients’ demographics and tumor characteristics. Percutaneous RFA was used to treat a

3092

C.-H. Liu et al.

TABLE 1 Patient demographics and tumor characteristics for 69 hepatic metastases

TABLE 2 Determinants of local progression by univariate analysis in 69 hepatic metastases

Characteristic

Value

Characteristic

Age (years)

65.13 ± 12.09 (42– 86)

Sex

Sex, M/F

36/33

Total no. of tumors of RFA treatment Tumor size (cm)

69 2.43 ± 1.23 (0.9–6.3)

RFA system (Covidien Inc./Boston Scientific)

45/24

No. of ablation per tumor

4.01 ± 1.58 (1–10)

Zone of ablation size (cm)

4.28 ± 1.71 (1.8– 10.8)

Post-ablation margin (cm)

0.99 ± 0.71 (0–4)

Local tumor progression rate after RFA

44.9 % (31/69)

Average local tumor progression time (month)

9.39 ± 6.08 (3–24)

Average follow-up time (month)

11.13 ± 7.79 (3–36)

Value

pa

Male

36

0.313

Female

33

Age (years) C67

35

\67

34

RFA system Covidien Inc. Boston Scientific

45

0.751

0.167

24

Tumor size (cm) C3.0

20

\3.0

49

0.024

Post-ablation margin (cm)

Data are presented as mean ± SD (range) or n RFA radiofrequency ablation

single tumor in 45 patients (83.3 %), 2 tumors in 5 patients (9.3 %), 3 tumors in 2 patients (3.7 %), and 4 tumors in 2 patients (3.7 %). Local tumor progression was identified in 31 (44.9 %) of 69 lesions during the follow-up period. When classified by tumor type, the colorectal hepatic metastases group had a higher local tumor progression rate than the non-colorectal hepatic metastases group, but the difference was not statistically significant (46.9 %, 15 of 32, vs. 43.2 %, 16 of 37; p = 0.812, Fisher’s exact test). Effects of Risk Factors on Local Progression Rate of Hepatic Metastases Table 2 shows the determinants of local tumor progression of hepatic metastases. Presence of a tumor size \3 cm was significantly related to local tumor progression of hepatic metastases. The cumulative rate of local tumor progression at 24 months was significantly higher for hepatic metastases with a tumor size of C3 cm (100 %) than for hepatic metastases with a tumor size of \3 cm (60.4 %) (Fig. 1a). Cox regression analysis showed that the tumor size of \3 cm was a borderline significant independent risk factor for local control [p = 0.055; relative risk 2.14; 95 % confidence interval (CI) 0.99–4.63]. Presence of a tumor contiguous with large vessels was significantly related to local tumor progression of hepatic metastases. The cumulative rate of local tumor progression at 24 months was significantly higher for hepatic metastases with a tumor contiguous with large vessels (93.8 %) than for hepatic metastases with a tumor not contiguous with large vessels (38.0 %) (Fig. 1b). Cox regression

C1.8 0.8

7

0.064

62

Contiguous with large vessels Yes

24

No

45

0.002

Subcapsular location Yes

37

No

32

0.134

RFA radiofrequency ablation a

Kaplan–Meier method and log-rank test

analysis showed that the absence of a tumor contiguous with large vessels was a significant independent risk factor for local control (p = 0.009; relative risk 2.65; 95 % CI 1.27–5.52). A post-ablation margin of\1.8 cm was not significantly related to local tumor progression of hepatic metastases. The cumulative rate of local tumor progression at 36 months was higher for hepatic metastases with a postablation margin of \1.8 cm (69.2 %) than for hepatic metastases with a post-ablation margin of C1.8 cm (52.4 %), but the difference was not statistically significant (p = 0.064, log-rank test) (Fig. 1c). Cox regression analysis showed that the threshold post-ablation margin of 1.8 cm was not a significant independent risk factor for local control (p = 0.114; relative risk = 0.37; 95 % CI 0.11–1.27). Presence of a tumor with subcapsular location did not correlate with local tumor progression of hepatic metastases. The cumulative rate of local tumor progression at 36 months was higher for hepatic metastases with a nonsubcapsular location (69.4 %) than for hepatic metastases with a subcapsular location (68.3 %), but the difference was not significant (p = 0.134, log-rank test) (Fig. 1d). Cox regression analysis showed that a tumor with a


(b)

1.0

1.0 Tumor size 3.0 cm (n=20), cumulative tumor progression rate=100%

Logrank Test, p=0.024

0.8

0.6

0.4

Tumor size < 3.0 cm (n=49), cumulative tumor progression rate=60.4%

0.2

Rate of Tumor Progression (%)


(a)

3093

0.0

Contiguous with large vessels: Yes (n=24), cumulative tumor progression rate=93.8% 0.8

0.6

0.4

Contiguous with large vessels: No (n=45), cumulative tumor progression rate=38.0%

0.2

0.0 0.00

10.00

20.00

30.00

40.00

0.00

10.00

20.00

30.00

40.00

Follow–Up Period (month)


(c)

(d) 1.0

0.8

1.0 Post-ablation margin ≥ 1.8 cm (n=7), cumulative tumor progression rate=52.4%


0.6

0.4 Post-ablation margin < 1.8 cm (n=62), cumulative tumor progression rate=69.2%

0.2

0.0




0.8


Sub capsular location (n=37), cumulative tumor progression rate=68.3%

0.6

0.4

Nonsub capsular location (n=32), cumulative tumor progression rate=69.4%

0.2

0.0 0.00

10.00

20.00

30.00

40.00


0.00

10.00

20.00

30.00

40.00


FIG. 1 Curves of local tumor progression showing log-rank test results, calculated by Kaplan–Meier method, according to risk factors of local tumor progression and time. a Presence of tumor C3.0 cm in size was associated with significantly high rate of local tumor progression in hepatic metastases (p = 0.024). b Presence of a tumor contiguous with large vessels was associated with significantly high

rate of local tumor progression in hepatic metastases (p = 0.002). c Presence of post-ablation margin of \1.8 cm was not associated with significantly high rate of local tumor progression in hepatic metastases (p = 0.064). d Presence of tumor with subcapsular location was not associated with significantly high rate of local tumor progression in hepatic metastases (p = 0.134)

subcapsular location was not a significant independent risk factor for local control (p = 0.134; relative risk = 1.77; 95 % CI 0.84–3.72).

confined to the liver with the potential for a cure.1,10–13 When surgical resection is not an option, local ablation therapies are often considered for the management of hepatic metastases.8,14,15 Percutaneous RFA is an accepted loco-regional treatment of a nonresectable hepatic tumor because of its effectiveness and safety.2,16 Although it is known that local tumor progression is more common with hepatic metastases than with hepatocellular carcinoma (HCC), the purpose of the present study was to attempt to identify the risk factors for local tumor progression in patients with hepatic metastases treated with percutaneous RFA.17–21 With percutaneous RFA treatment of hepatic tumors, the tumor size appears to influence local tumor progression. An et al.22 reported that RFA is an effective procedure for the treatment of small hepatic metastases (B2.5 cm) from gastric adenocarcinoma, with a complete necrosis rate of [90 %. We found that local control of hepatic metastases \3 cm in diameter was significantly more successful than

Complications There were no treatment-related deaths or major complications. Ten minor complications were observed (18.5 %, 10 of 54). These included small asymptomatic pneumothoraces (n = 5), small perihepatic hematomas (n = 2), a relatively small pleural effusion (n = 2), and a small peri-hepatic biloma (n = 1). All minor complications were managed conservatively. DISCUSSION The liver is second only to lymph node as the most common site of metastasis from solid tumors.1,9 Metastasectomy is the standard treatment when the disease is

3094

that of hepatic metastases C3 cm in diameter (p = 0.024, log-rank test). Cox regression analysis also showed that the tumor size \3 cm was a significant independent risk factor for local control. The ablation volume of RFA for hepatic tumors has increased significantly in the past several years owing to refinements of the ablation technique as well as the introduction of more powerful generators.19,20,23 Nonetheless, ablation of a large tumor still requires multiple overlapping ablation zones to achieve an adequate ablation margin in three dimensions. In a study by Liu et al.,8 55 patients with 69 HCCs underwent percutaneous RFA; a 0.4-cm postablation margin was adequate to control local tumor progression of HCC \2.5 cm, with no local tumor progression. To our knowledge, the optimal post-ablation margin has yet to be determined for hepatic metastases. Cady et al. 24 recommended a surgical standard of a 2-cm margin but settled for a 1-cm hepatic resection margin for colorectal hepatic metastases to prevent local progression. In our study, the local tumor progression rate was lower in hepatic metastases with a post-ablation margin C1.8 cm (42.9 %, 3 of 7) compared with a post-ablation margin \1.8 cm (45.2 %, 28 of 62). However, our data showed that the 1.8-cm post-ablation margin for hepatic metastases was not a significant independent risk factor for local control. Another factor that may affect the success rate of percutaneous RFA is the tumor location. For example, a tumor may be contiguous with a large vessel or a subcapsular location. In the case of an adjacent large intrahepatic vessel, an adequate ablation margin is difficult to achieve because of heat-sink effects of large vessels. In the present study, the size of post-ablation margin of hepatic metastases not contiguous with large vessels was larger than the size of post-ablation margin of hepatic metastases contiguous with large vessels, but the difference was not statistically significant (p = 0.196, Student’s t test). Nakazawa et al.25 used percutaneous RFA for the treatment of small HCC (B3 cm) and showed a 25.6 % rate (11 of 43) of local tumor progression in a series of 85 patients with HCC and with a tumor located within 5 mm of an intrahepatic vessel; this rate was 7.1 % (3 of 42) for tumors not contiguous with vessels. In the present study, local control of hepatic metastases when a tumor is not contiguous with a large intrahepatic vessel (diameter [3 mm) was significantly better than hepatic metastases with a tumor contiguous with a large vessel (p = 0.002, log-rank test). Cox regression analysis also showed that the tumor contiguous with large vessels was a significant independent risk factor for local control. Our results showed that in cases of hepatic metastases with subcapsular location, the local tumor progression after percutaneous RFA (48.6 %, 18 of 37) was not significantly

C.-H. Liu et al.

different from that of tumors with non-subcapsular location (40.6 %, 13 of 32). These results are in line with the observation by Sartori et al.16 They used ultrasound-guided percutaneous RFA for the treatment of 178 hepatic metastases. The local tumor progression was observed in 6 of 44 nodules with subcapsular location and 13 of 134 nodules with non-subcapsular location; the difference was not significant (p = 0.651). Some authors have reported that tumors with a subcapsular location and tumors abutting hollow viscera have a higher rate of local tumor progression and increased risk of complications.18,26 However, our results showed that the local tumor progression rate in metastatic nodules with subcapsular location was not significantly different from that of metastatic nodules with non-subcapsular location. In the present work, the overall minor complication rate was 18.5 % (10 of 54), and there was no significant difference between the nodules with subcapsular location (20.0 %, 6 of 30) and nodules with non-subcapsular location (16.7 %, 4 of 24). Tung-Ping Poon et al.27 reported that HCCs with subcapsular location are not a risk factor of recurrence after surgical resection. Kim et al.28 also reported that local tumor progression at the subcapsular ablative margin was undetectable after percutaneous RFA of HCCs with subcapsular location. In the present study, most patients underwent percutaneous RFA under CT guidance (74.1 %, 40 of 54) to place the electrode closer to the liver capsule (within l cm). The use of a straight internally cooled single electrode (64.8 %, 35 of 54) made it easier to monitor the location of the electrode tip under CT guidance during percutaneous RFA. This may explain in part why local tumor progression was not significantly different between metastatic nodules with subcapsular and non-subcapsular locations. This study has some limitations. First, there was small number of patients and tumors, which limits the power of the study. Second, we included 12 types of hepatic metastases, and the different biological behaviors of the different primary tumors may affect the outcomes. Third, the method of determining post-ablation margin was based on the zone of ablation on the 1-month follow-up image study. However, the post-ablation margin was not always circumferential or symmetric around the tumor, possibly leading to bias. Finally, overall survival was not included in the outcome measures because the purpose of our study was to evaluate the factors influencing local tumor progression after percutaneous RFA of hepatic metastases. In conclusion, the present results indicate that percutaneous RFA is a safe and effective treatment for hepatic metastases. Our data suggest that a tumor size of C3 cm and the presence of a contiguous large vessel are associated with high rates of local tumor progression. Percutaneous RFA with multiple electrodes approaches or microwave


ablation with multiple antenna insertions is an alternative technique to achieve a greater ablation zone in cases of high rate of local tumor progression for hepatic metastases, which should be explored in further studies. ACKNOWLEDGMENT Supported in part by a grant from TriService General Hospital (TSGH C99-052). The Institutional Review Board for Human Investigation of the Tri-Service General Hospital, National Defense Medical Center (TSGHIRB 098-05-260), approved this study. DISCLOSURE

The authors declare no conflict of interest.

REFERENCES 1. Chen J, Tang Z, Dong X, et al. Radiofrequency ablation for liver metastasis from gastric cancer. Eur J Surg Oncol. 2013;39:701–6. 2. Park SY, Kim JH, Won HJ, Shin YM, Kim PN. Radiofrequency ablation of hepatic metastases after curative resection of extrahepatic cholangiocarcinoma. AJR Am J Roentgenol. 2011;197:1129–34. 3. Namasivayam S, Martin DR, Saini S. Imaging of liver metastases: MRI. Cancer Imaging. 2007;7:2–9. 4. Chiou YY, Chou YH, Chiang JH, Wang HK, Chang CY. Percutaneous ultrasound-guided radiofrequency ablation of colorectal liver metastases. Chin J Radiol. 2005;30:153–8. 5. Tateishi R, Shiina S, Teratani T, et al. Percutaneous radiofrequency ablation of hepatocellular carcinoma. an analysis of 1000 cases. Cancer. 2005;103:1201–9. 6. Shibata T, Shibata T, Maetani Y, Isoda H, Hiraoka M. Radiofrequency ablation for small hepatocellular carcinoma: prospective comparison of internally cooled electrode and expandable electrode. Radiology. 2006;238:346–53. 7. Livraghi T, Solbiati L, Meloni F, Ierace T, Goldberg SN, Gazelle GS. Percutaneous radiofrequency ablation of liver metastases in potential candidates for resection: the ‘‘test-of-time’’ approach. Cancer. 2003;97:3027–35. 8. Liu CH, Arellano RS, Uppot RN, Samir AE, Gervais DA, Mueller PR. Radiofrequency ablation of hepatic tumours: effect of post-ablation margin on local tumour progression. Eur Radiol. 2010;20:877–85. 9. Weiss L, Grundmann E, Torhorst J, et al. Haematogenous metastatic patterns in colonic carcinoma: an analysis of 1541 necropsies. J Pathol. 1986;150:195–203. 10. Curley SA, Izzo F, Deirio P, et al. Radiofrequency ablation of unresectable primary and metastatic hepatic malignancies: results in 123 patients. Ann Surg. 1999;230:1–8. 11. Ksienski D, Woods R, Speers C, Kennecke H. Patterns of referral and resection among patients with liver-only metastatic colorectal cancer (MCRC). Ann Surg Oncol. 2010;17:3085–93. 12. Pockaj BA, Wasif N, Dueck AC, et al. Matastasectomy and surgical resection of the primary tumor in patients with stage IV breast cancer: time for a second look? Ann Surg Oncol. 2010;17:2419–26.

3095 13. Saxena A, Chua TC, Sarkar A, et al. Progression and survival after radical hepatic matastasectomy of indolent advance neuroendocrine neoplasms (NENs) supports an aggressive surgical approach. Surgery. 2011;149:209–20. 14. Mazzaferro V, Battiston C, Perrone S, et al. Radiofrequency ablation of small hepatocellular carcinoma in cirrhotic patients awaiting liver transplantation: a prospective study. Ann Surg. 2004;240:900–9. 15. Gillams AR, Lees WR. Radiofrequency ablation of colorectal liver metastases. Abdom Imaging. 2005;30:419–26. 16. Sartori S, Tombesi P, Macario F, et al. Subcapsular liver tumors treated with percutaneous radiofrequency ablation: a prospective comparison with nonsubcapsular liver tumors for safety and effectiveness. Radiology. 2008;248:670–9. 17. Gervais DA, Arellano RS, Mueller PR. Percutaneous radiofrequency ablation of ovarian cancer metastasis to the liver: indications, outcomes, and role in patient management. AJR Am J Roentgenol. 2006;187:746–50. 18. Komorizono Y, Oketani M, Sako K, et al. Risk factors for local recurrence of small hepatocellular carcinoma tumors after a single session, single application of percutaneous radiofrequency ablation. Cancer. 2003;97:1253–62. 19. Montgomery RS, Rahal A, Dodd GD 3rd, Leyendecker JR, Hubbard LG. Radiofrequency ablation of hepatic tumors: variability of lesion size using a single ablation device. AJR Am J Roentgenol. 2004;182:657–61. 20. Hansler J, Frieser M, Tietz V, et al. Percutaneous radiofrequency ablation of liver tumors using multiple saline-perfused electrodes. J Vasc Interv Radiol. 2007;18:405–10. 21. Harrison LE, Koneru B, Baramipour P, et al. Locoregional recurrences are frequent after radiofrequency ablation for hepatocellular carcinoma. J Am Coll Surg. 2003;197:759–64. 22. An JY, Kim JY, Choi MG, et al. Radiofrequency ablation for hepatic metastasis from gastric adenocarcinoma. Yonsei Med J. 2008;49:1046–51. 23. Ng KK, Poon RT. Radiofrequency ablation for malignant liver tumor. Surg Oncol. 2005;14:41–52. 24. Cady B, Jenkins RL, Steele GD, et al. Surgical margin in hepatic resection for colorectal metastasis: a critical and improvable determinant of outcome. Ann Surg. 1998;227:566–71. 25. Nakazawa T, Kokubu S, Shibuya A, et al. Radiofrequency ablation of hepatocellular carcinoma: correlation between local tumor progression after ablation and ablative margin. AJR Am J Roentgenol. 2007;188:480–8. 26. Hori T, Nagata K, Hasuike S, et al. Risk factors for local recurrence of hepatocellular carcinoma tumors after a single session of percutaneous radiofrequency ablation. J Gastroenterol. 2003;38: 977–81. 27. Tung-Ping Poon R, Fan ST, Wong J. Risk factors, prevention, and management of postoperative recurrence after resection of hepatocellular carcinoma. Ann Surg. 2000;232:10–24. 28. Kim YS, Lee WJ, Rhim H, Lim HK, Choi D, Lee JY. The minimal ablative margin of radiofrequency ablation of hepatocellular carcinoma ([2 and\5 cm) needed to prevent local tumor progression: 3D quantitative assessment using CT image fusion. AJR Am J Roentgenol. 2010;195:758–65.