The role of contrast-enhanced ultrasound in planning treatment protocols for hepatocellular carcinoma before radiofrequency ablation
M.H. Chen,W. Yang,K. Yan,Y. Dai,W. Wu,Z.H. Fan,M.R. Callstrom,J.W. Charboneau
DOI: https://doi.org/10.1016/j.crad.2006.12.013
IF: 3.389
2007-01-01
Clinical Radiology
Abstract:Materials and methods From 2002 to 2005, 81 patients with 110 HCCs underwent CEUS with SonoVue before RFA treatment (group A). Eighty six patients with 112 HCCs who underwent US without contrast enhancement before RFA served as the control group (group B). The average diameters of the lesions in group A and group B were 3.6 ± 1.1 cm and 3.5 ± 1.1 cm, respectively. There were no significant differences in clinical data between the two groups. Regular follow-up after treatment was performed using contrast-enhanced computed tomography (CECT). After treatment, complete necrosis was defined as the absence of viable tissue in treated tumours at the 1-year follow-up CECT. Results Using CEUS an additional seven small lesions (≤2.0 cm) were found compared with those found using CECT and conventional US. CEUS showed that 56.4% of lesions (62/110 tumours) were larger in size and 49.1% (54/110 tumours) became more irregular in shape during the arterial phase than on conventional US. Feeding vessels were detected using CEUS in 52 (91.2%) of 57 lesions that were larger than 3.5 cm. The follow-up period was at least 1 year for each case. The complete tumour necrosis rate in group A was significantly higher than that in group B (92.2% versus 83.0%; p = 0.036). Conclusion CEUS can be used to more accurately define the size and contour of lesions, and to detect additional small or satellite lesions and the feeding vessel of HCC tumours. CEUS provided important information for designing the ablation protocol, and might improve the efficacy of RFA. Introduction The clinical value of contrast-enhanced ultrasound (CEUS) used for detection and characterization of focal liver lesions has been demonstrated by several clinical studies. 1–5 Additionally, CEUS has been shown to be of value in the assessment of patients who have received local therapy for hepatocellular carcinoma (HCC), particularly RFA. 6–8 However, there are few reports on the value of CEUS before RFA treatment. Minami et al. 9 reported their experience using CEUS for lesions with an indistinct border, indicating that CEUS can improve the definition of the target lesion. The present study was undertaken to assess the value of CEUS in planning RFA treatment strategies and to compare the efficacy of RFA after US. Materials and methods Patients diagnosed with HCC were recruited to the study if they met the following criteria: (1) no more than four lesions (not including satellite lesions); (2) maximum diameter of main tumour less than 8 cm; (3) absence of thrombosis in the main branch of the portal vein and extrahepatic metastases; (4) prothrombin time ratio greater than 50% (normal range: 70–130%); and platelet count greater than 50,000/μl (normal range: 100,000–300,000/μl); (5) proof of malignancy for at least one hepatic lesion was obtained using US-guided fine-needle biopsy. The study was performed with the approval of the institutional ethics committees. Written informed consent was obtained from all patients after the nature of the procedures was fully explained. Patients From July 2002 to March 2005, 167 HCC patients met the inclusion criteria and underwent US-guided percutaneous RFA treatment. These patients were alternately apportioned to two groups: 81 patients received RFA within 10 min after a CEUS examination (group A); during the same period, another 86 HCC patients received US before RFA treatment and served as the control group (group B). Group A included 68 men and 13 women (age range 35–80 years; mean age 60.8 ± 11.4 years). The maximum diameter of the tumours ranged from 0.7–7.6 cm (mean, 3.6 ± 1.1 cm) on conventional US. Fifty-nine patients had AJCC stage II–III HCC. The liver function status of 58 patients was Child–Pugh class A, and in 23 it was class B. Seventy-two (88.9%) patients had liver cirrhosis. Thirty-eight (46.9%) patients had elevated alpha fetoprotein (AFP). Group B included 72 men and 14 women (age range 24–77 years; mean age, 59.6 ± 10.5 years). The maximal diameter of the tumours ranged from 1.2–7.8 cm (3.5 ± 1.1 cm) on conventional US. Sixty-two patients had AJCC stage II–III HCC. The liver function status of 60 patients was Child–Pugh class A, and in 26 it was class B. Seventy-eight (90.7%) patients had liver cirrhosis. Thirty-nine (45.3%) patients had elevated AFP. No statistically significant differences between the group A and group B were observed with respect to baseline characteristics ( Table 1 ). Examination method and equipment First, US was used to determine the location, size, number, and border of the tumour, then the blood supply and feeding vessels were examined using colour US in the two groups. The US images were recorded on hard copy. Localization of the tumour with indistinct margins on US required CECT as a reference. For group A, the contrast perfusion pattern of tumour in each phase was recorded and fan-shaped imaging was used for the neighbouring section by slightly leaning the probe to observe the whole tumour and surrounding area. After adequate diagnostic information of the key lesion was acquired, the whole liver was examined to find any other abnormal wash-out lesions. A second injection of contrast medium was given about 8 min later when the first injection was not satisfactory or if another lesion needed to be observed. The CEUS procedure was recorded on VHS videotape. Two radiologists (each having more than 10 years experience in interventional US and more than 50 cases experience in CEUS) reviewed the recorded images and recorded information regarding tumour characteristics, such as the number of tumours, location, morphology, maximum diameter, border, satellite lesions, feeding vessel, relationship with surrounding important structures (such as intestine, diaphragm, or a large vessel). The RFA protocol was designed on the basis of CEUS information in group A, whereas it was designed based on the US information in group B. CECT within 1 month before RFA provided reference information for the RFA protocol in both groups, because all RFA procedures were guided by ultrasound. Low-mechanical index (MI) and real-time contrast-enhanced harmonic ultrasound with SonoVue (Bracco SpA, Italy) was performed for group A. SonoVue suspension (2.4 ml) was administered through a cubital vein by bolus injection (in 1–3 s). The Technos DU8 (Esaote, Italy), IU22 (Philips Medical Systems, Netherlands), and LOGIQ 9 (General electric, MILWAUKEE, WI, America) were adapted for this study. The probe frequency was approximately 2.5–5 MHz. The MI used in CEUS imaging was 0.04–0.1. CT examinations were performed with Plus 4 (Siemens, Germany) with 5-mm collimation and table speed 7.5-mm/s. A total of 100 ml non-ionic contrast material (300 mg iodine/ml, Omnipaque, Amersham, Shanghai, China) was administered at a rate of 3 ml/s with a power injector (OP 100, Medrad, Pittsburgh, PA, USA). Images were acquired before and at 25 and 60 s following intravenous contrast material injection, during the hepatic arterial and portal venous phases, respectively. RFA treatment The same two radiologists performed all percutaneous RFA treatments. First, the number of tumours, location, size, and shape were identified. A safety margin to cover the tumour and at least 0.5 cm of the surrounding tissue was planned. Tumours larger than 3.5 cm in diameter were ablated with multiple overlapping spheres, and the amount and site of the ablation spheres were determined according to tumour size and shape. 10,11 The ablation sequence was determined on an individual basis for each case; the area to be ablated first was the portion where a feeding vessel entered the tumour or the portion close to major structures. 11 Real-time CEUS-guided RFA was performed for small lesions that were only seen on CEUS. Furthermore, high mechanical index post-contrast ultrasonography improved the display of HCC and provided reference information for localization. 12 The radiofrequency system used in this study was a 460-kHz generator unit (Model 1500; RITA Medical System, Mountain View, CA, USA). Expandable electrodes consisted of an outer insulated needle having a calibre of 14 G and a length of 15 cm. A sphere-like coagulation area of 3–5 cm in diameter could be produced when the electrode was inserted into the tumour and nine prongs were deployed from the cannula. All patients underwent percutaneous RFA under moderate sedation anaesthesia, induced with intravenous administration of 2.5–5 mg midazolam (Roche, Basel, Switzerland) and 50–100 μg fentanyl (Fentaini; Renfu, Yichang, China). Intravenous infusion of propofol (Diprivan; Zeneca, Macclesfield, UK; 1–2 mg/kg) was used to temporarily enhance anaesthesia. Efficacy assessment and follow-up CECT was performed in all patients 1 month after RFA, together with laboratory tests and sonograms. Magnetic resonance imaging was used in two patients who were allergic to iodinated contrast media. Imaging and laboratory follow-ups were conducted every 3 months in the first year, and every 4–6 months in the following years. RFA was considered successful if no arterial and portal enhancement was seen in and around the tumour. If a recurrent or new lesion was found, ultrasound-guided biopsy was performed to confirm its pathology. The follow-up period was at least 1 year for each case. Complete necrosis was defined as the absence of viable tissue in the treated tumour at the 1 year CECT follow-up examination. Two radiologists who had at least 10 years experience in liver CT reviewed the CT results to assess RFA efficacy with consensus. Statistic analysis In the analysis of patient characteristics, data are expressed as mean ± SD. Differences in proportions between groups were analysed using the chi-square test. Parametric data were compared using the Student's t -test, and nonparametric data were compared using the Mann–Whitney U test. All p -values were derived from two-tailed tests, and a level of less than 0.05 was accepted as statistically significant. SPSS statistical analysis software (SPSS, Chicago, IL, USA) was used. Results On average 3.1 ml (range 2.4–4.8 ml) SonoVue was used in group A before RFA. The detection rates of US, CEUS, and CECT examination are summarized in Table 2 . At CEUS, an additional seven small lesions (≤2 cm) were found that were not detected by CECT and fundamental US. When compared with US, CEUS detected 12 additional small HCC. Of these, four were new HCC in the liver separate from the original lesion, and the other eight were minute satellite lesions around the main tumour. When compared with CECT, CEUS found seven additional small HCC ( Fig. 1 ). Of these, two were new HCC (≤1 cm) and the other five were minute satellite lesions. These lesions were diagnosed as HCC according to the typical HCC perfusion mode of fast wash-in at arterial phase and wash-out at parachymal phase, and minute lesions were confirmed by CEUS-guided biopsy. 13 However, two small lesions under the diaphragm detected by CECT were not detected using CEUS or US. In group A, 91.8% of the lesions (101 of 110 lesions) were hypervascular in the arterial phase and washed out in the parenchymal phase, and therefore, likely to represent HCC. The Technos DU8 and LOGIQ 9 were used in 65 cases, IU22 in 16 cases. All of these US machines could sensitively demonstrate the hypervascularity of HCC, as well as detect satellite lesions and feeding vessels, and also guide RFA treatment. Only 47 lesions (42.7%) showed a distinct margin on US, although 98 lesions (89.1%) had a distinct margin in the arterial phase of CEUS ( p < 0.001). Of the 110 HCC lesions, 62 (56.4%) became larger in size and 54 (49.1%) became more irregular in shape than could be demonstrated on US ( Figs. 2–4 ). Among the 62 lesions that became larger on CEUS, 44 (70.9%) had an indistinct margin on US, and 42 (77.8%) of the 54 lesions that became more irregular on CEUS had an indistinct margin on US. The relationship between tumour margin delineation on US and CEUS arterial phase behaviour is listed in Table 3 . Regarding tumour size measurements, in 63 lesions with indistinct margins, the median maximum diameter was 2.8 cm on fundamental US and significantly increased to 3.5 cm on CEUS. The median difference was 0.3 cm and the range was from 0.4–2.4 cm. This difference was statistically significant ( p < 0.001; see Fig. 2 ). Non-contrast colour US can give useful information with respect to feeding vessels and vascularity of the tumour, but it had lower sensitivity than CEUS. For the 57 lesions with diameters of more than 3.5 cm in group A, the rich blood flow of the inner tumour or of the peripheral tumour was observed for 38 lesions (66.7%), and the main feeding vessel could be seen in only 18 lesions (31.6%) with colour US. At CEUS, the main feeding vessel was observed for 52 lesions (91.2%) during the arterial phase ( p < 0.001). The treatment strategy for lesions with feeding vessels was first to ablate the entrance area of the feeding vessel, and then use repeated small ablations to enhance the coagulation effect in this area. In this way, 48 (92.3%) of the 52 tumours with feeding vessel in group A achieved complete necrosis. Eighty-one patients with 115 HCCs (including additional five lesions found using CEUS examination) in group A and 86 patients with 112 HCCs in Group B underwent RFA. The average number of ablations in initial RFA was 2.8 for group A and 2.5 for group B ( p > 0.05). In tumours larger than 3.5 cm, the number was 4.8 and 4.1 for group A and group B respectively ( p = 0.032). On the basis of the follow-up at least 1 year after RFA, the complete necrosis rate after initial RFA in group A was 92.2% (106/115 tumours), which was significantly higher than 83% (93/112 tumours) of group B ( p = 0.036). In group A, the AFP values decreased to a normal level in 73.6% (28/38) of patients within 1 month, and in six patients within 3 months. In group B, AFP values decreased to a normal level in 66.7% (26/39) of patients within 1 month, and in five patients within 3 months. The overall complication rate in the present study was 2.4% (four of 167 cases). The complication rate in group A was lower than in group B (1.2% vs. 3.5%), but there was no significant difference. Needle tract seeding was found 8 months after RFA treatment in one case of group A and, 12 months after RFA in one case of group B. Peritoneal haemorrhage occurred during RFA treatment in two cases in group B and was controlled by haemostatic drugs. Discussion The success rate of RFA treatment for liver tumours is influenced by tumour size, border and invasive area, tumour staging, and treatment protocol. 14–16 Many studies have reported that enlarging the ablation area with a safety margin of 0.5–1 cm could decrease the risk of tumour recurrence. 17,18 The mathematical protocol for multiple ablations of large tumours previously attracted attention. 10,19–21 Recently, on the basis of the experience at our hospital, even though the ablation area was enlarged according to the mathematical protocol, some patients still had local tumour progression within 3–6 months after the treatment. The main reasons for local treatment failure could be the following: the invasive area was not identified correctly or small satellite lesions were missed; the tumour had a rich blood supply and the heat sink effect caused incomplete ablation. Thus, an effective imaging method is needed to accurately identify the tumour extent, number of tumours, and feeding vessels before local treatment. Solbiati et al. 22 first reported their experience in using CEUS before, during, and after RFA treatment, and their results indicated that CEUS represented a significant improvement in the detection of lesions, selection of patients for treatment, and all steps of tumour ablative treatment. They emphasized the role of CEUS in the differentiation of necrotic changes from viable tumour. Minami et al. 9 indicated that CEUS helps in the placement of RFA electrodes in HCC that cannot be adequately depicted by US. In the present study, the role of CEUS before RFA was further investigated, and we showed that HCC with indistinct margins usually appear larger or irregular on CEUS than on US. This finding suggests that CEUS might demonstrate more accurately the extent of HCC to more accurately guide RFA. Our results also showed that CEUS was useful in detecting satellite lesions, feeding vessels, and displaying the relationship between the tumour and the major organs. Thus, CEUS could provide valuable clinical information for planning treatment protocols and improving the treatment success rate. The results of the present study showed that CEUS was useful in differentiating regenerative nodules of cirrhotic livers from small HCC. Furthermore, CEUS has a high sensitivity for detecting small satellite lesions around the main tumour in cirrhotic livers because the lesions have excellent hyperechoic contrast in the arterial phase and/or are hypoechoic in the parenchymal phase. Maruyama et al. 23 investigated the efficacy of CEUS in detecting liver tumours in 15 rabbits with VX-2 tumours and found that CEUS was superior to non-CEUS in delineating the nodules and in detecting small daughter nodules. In this study, four small tumours and eight minute satellite lesions that were missed using US were identified and treated, together with the main tumours, by repeat ablation or extending the ablation range. However, tumour location limited the visualization of lesions by CEUS. In this study, two small lesions under the diaphragm were not found using CEUS but were identified using CECT. Seven small HCCs, which consisted of two tiny lesions (≤1 cm) and five minute satellite lesions, were detected using CEUS but not using CECT. The combination of CEUS with CECT was important for accurately identifying small lesions before RFA treatment and helped to decrease local progression after treatment. Correct identification of the exact size and extent of HCC before RFA are important for designing treatment protocols. 24 However, US often produces errors in tumour measurement and the definition of tumour margins. Large infiltrative HCCs have an indistinct margin and no capsule. US can only depict the approximate shape of the hyperechoic or hypoechoic tumour area, but CEUS provides functional, as well as anatomical, information about blood flow in the tumour, and is especially useful in detecting abnormal blood vessel architecture that surrounds a tumour. Livraghi et al. 14 reported the RFA results of medium and large HCC, and found that there was a distinct difference in tumour necrosis rate between invasive tumours and non-invasive tumours. One reason for this might be that conventional US could not accurately measure the size of invasive tumours before treatment, resulting in insufficient ablation. Thus, it is crucial to define the exact morphological features of the liver tumour before RFA treatment. On CEUS, the enhancement area of 56.4% lesions in the arterial phase was bigger than that on US examination. This finding was consistent with the results reported by Catalano et al., 25 who found that 47% of lesions were larger on CECT examination compared with plain CT. In this study, of the lesions with indistinct margins on US, 71% were larger and had better defined margins on CEUS. However, 62% of the tumours with distinct margins or capsules showed no observed changes in size after contrast agent administration. The same phenomenon was seen where lesions were partly ill-defined, and partly capsulated (see Fig. 3 ). Where tumours became larger and more irregular in the arterial phase, CEUS helped to design a suitable ablation range. The potential mechanism might be related to tumour vessel structure, with CEUS demonstrating the tumour extent by perfusion of the microvessels invading the surrounding tissue, thus helping to delineate it (see Figs. 3 and 4 ). However, the exact mechanism requires further study. Feeding vessels could be identified more sensitively with CEUS than with colour US, and microbubbles could be observed floating from the feeding vessel into the tumour during the early arterial phase. Perfusion was sometimes so fast that the image of the feeding vessel was not easily captured. The imaging plane for CEUS was selected using colour US, and patients were asked to hold their breath during the early arterial phase. In this way, the main feeding vessel was seen and sparkling hyperechoic spots were observed flowing rapidly to the tumour and surrounding it. For 52 lesions (91.2%) larger than 3.5 cm, the feeding vessel was depicted clearly in the arterial phase. When a tumour feeding vessel was visible, the entrance area where it entered the tumour was ablated first. If there was still blood flow in the ablation region after one ablation, additional small ablations were performed to enhance the coagulation effect in this area. CEUS was also helpful for demonstrating the relationship of the tumour to the surrounding structures, such as intestine, diaphragm, and gallbladder as the tumour border usually became clearer with CEUS. This information helped reduce potential complications. In this study, the complication rate in group A (1.2%) was lower than that in group B (3.5%). The remaining liver parenchyma is examined after the arterial phase to search for any abnormal wash-out lesions, but some small HCC with arterial enhancement without wash-out might be not found. Cases were selected with less than five lesions based on CECT and US before CEUS examination, so the study group did not represent the entire HCC population. This might decrease the detection rate of CECT for small lesions. Furthermore, the long-term outcome, i.e., survival between the two groups, was not studied as the follow-up period was not long enough. In conclusion, in this study the complete necrosis rate after RFA treatment in group A was higher than that in group B (92.2% versus 83.0%; p = 0.036), which might be due to the high sensitivity of CEUS for identifying small lesions, tumour extent, and feeding vessels. In our experience, the use of CEUS provided valuable information for guiding RFA treatment of HCC. References 1 M.H. Chen Y. Dai K. Yan The role of contrast-enhanced ultrasound on the diagnosis of small hepatocellular carcinoma (≤3 cm) in patients with cirrhosis Hepatol Res 35 2006 281 288 2 C. Nicolau V. Catala R. Vilana Evaluation of hepatocellular carcinoma using SonoVue, a second generation ultrasound contrast agent: correlation with cellular differentiation Eur Radiol 14 2004 1092 1099 3 E. Leen W.J. Angerson S. Yarmenitis Multi-centre clinical study evaluating the efficacy of SonoVue (BR1), a new ultrasound contrast agent in Doppler investigation of focal hepatic lesions Eur J Radiol 41 2002 200 206 4 E. Quaia F. Calliada M. Bertolotto Characterization of focal liver lesions with contrast-specific US modes and a sulfur hexafluoride-filled microbubble contrast agent: diagnostic performance and confidence Radiology 232 2004 420 430 5 A. Von Herbay C. Vogt R. Willers Real-time imaging with the sonographic contrast agent SonoVue: differentiation between benign and malignant hepatic lesions J Ultrasound Med 23 2004 1557 1568 6 L. Solbiati T. Ierace M. Tonolini Guidance and monitoring of radiofrequency liver tumor ablation with contrast-enhanced ultrasound Eur J Radiol 51 Suppl 2004 S19 S23 7 L. Solbiati M. Tonolini L. Cova Monitoring RF ablation Eur Radiol 14 Suppl. 8 2004 P34 P42 8 M.H. Chen K. Yan Y. Dai New gray scale contrast ultrasound technique in liver neoplasms diagnosis and radiofrequency ablation application Chin J Med Imaging Technol 20 2004 326 330 9 Y. Minami M. Kudo T. Kawasaki Treatment of hepatocellular carcinoma with percutaneous radiofrequency ablation: usefulness of contrast harmonic sonography for lesions poorly defined with B-mode sonography AJR Am J Roentgenol 183 2004 153 156 10 M.H. Chen W. Yang K. Yan Protocol for radiofrequency ablation of large liver tumors and its clinical application in 110 patients Radiology 232 2004 260 271 11 M.H. Chen W. Yang K. Yan Treatment strategy to optimize radiofrequency ablation for liver malignancies J Vasc Interv Radiol 17 2006 671 683 12 M.H. Chen W. Yang Y. Dai High mechanical index post-contrast ultrasonography improves tissue structural display of hepatocellular carcinoma Chin Med J (Engl) 118 2005 2046 2051 13 Wu Wei M.H. Chen S.S. Yin The role of contrast-enhanced sonography of focal lesion before percutaneous biopsy AJR Am J Roentgenol 187 2006 752 761 14 K. Kosari M. Gomes D. Hunter Local, intrahepatic, and systemic recurrence patterns after radiofrequency ablation of hepatic malignancies J Gastrointest Surg 6 2002 255 263 15 R.J. Bleicher D. Allegra D.T. Nora Radiofrequency ablation in 447 complex unresectable liver tumors: lessons learned Ann Surg Oncol 10 2003 52 58 16 T. Livraghi F. Meloni S.N. Goldberg Hepatocellular carcinoma: radio-frequency ablation of medium and large lesions Radiology 214 2000 761 768 17 G.S. Gazelle S.N. Goldberg L. Solbiati Tumor ablation with radio-frequency ablation energy Radiology 217 2000 633 646 18 B. Cady R.L. Jenkins G.D. Steele Jr. Surgical margin in hepatic resection for colorectal metastases: a critical and improvable determinant of outcome Ann Surg 227 1998 566 571 19 J.P. McGahan G.D. Dodd III Radiofrequency ablation of the liver:current status AJR Am J Roentgenol 176 2001 3 16 20 Y.S. Khajanchee D. Streeter L.L. Swanstrom A mathematical model for preoperative planning of radiofrequency ablation of hepatic tumors Surg Endosc 18 2004 696 701 21 G.D. Dodd III M.S. Frank M. Aribandi Radiofrequency thermal ablation: computer analysis of the size of the thermal injury created by overlapping ablations AJR Am J Roentgenol 177 2001 777 782 22 T. Solbiati T. Ierace M. Tonolini Guidance and control of percutaneous treatments with contrast-enhanced ultrasound Eur Radiol 13 2003 N87 N90 23 H. Maruyama S. Matsutani H. Saisho Real-time blood-pool images of contrast enhanced ultrasound with definity in the detection of tumor nodules in the liver Br J Radiol 78 2005 512 518 24 M.H. Chen W. Yang K. Yan Mathematical protocol for radiofrequency ablation of liver tumors and its clinical application Natl Med J China 84 2004 203 208 25 O. Catalano M. Esposito A. Nunziata Multiphase helical CT findings after percutaneous ablation procedure for hepatocellular carcinoma Abdom Imaging 25 2000 607 614