- Email: [email protected]
Role of Sonazoid-enhanced three-dimensional ultrasonography in the evaluation of percutaneous radiofrequency ablation of hepatocellular carcinoma
European Journal of Radiology 75 (2010) 91–97
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Role of Sonazoid-enhanced three-dimensional ultrasonography in the evaluation of percutaneous radiofrequency ablation of hepatocellular carcinoma Wen Luo a,b , Kazushi Numata a,∗ , Manabu Morimoto a , Takashi Oshima a , Michio Ueda a , Masahiro Okada c , Shigeo Takebayashi d , Xiaodong Zhou b , Katsuaki Tanaka a a
Gastroenterological Center, Yokohama City University Medical Center, 4-57 Urafune-cho, Minami-ku, Yokohama, Kanagawa, 232-0024, Japan Department of Ultrasound, Xijing Hospital, Fourth Military Medical University, 15th Changle Xi Road, Xi’an, ShaanXi, 710032, China c Department of Radiology, Kinki University School of Medicine, 377-2 Ohno-Higashi, Osaka-Sayama, Osaka, 589-8511, Japan d Department of Radiology, Yokohama City University Medical Center, 4-57 Urafune-cho, Minami-ku, Yokohama, Kanagawa, 232-0024, Japan b
a r t i c l e
i n f o
Article history: Received 23 January 2009 Accepted 13 March 2009 Keywords: Three-dimensional ultrasonography Three-dimensional computed tomography Contrast agent Radiofrequency ablation Hepatocellular carcinoma
a b s t r a c t Objective: We investigated contrast-enhanced three-dimensional ultrasonography (CE 3D US) with contrast agent Sonazoid for evaluating the effect of percutaneous radiofrequency (RF) ablation of hepatocellular carcinomas (HCCs). Methods: 63 HCCs were treated by US-guided percutaneous RF ablation. CE 3D US after bolus injection of 0.2 mL of Sonazoid was performed 5–7 days before and 1 day after RF ablation. CE 3D computed tomography (CT) was performed 5–7 days before and 1 month after the ablation, and during the followup period. Multiplanar images in three orthogonal planes and US/CT angiograms were reconstructed on both modalities. Two blinded observers reviewed the images on both modalities to evaluate the ablation effects. Results: After RF ablation, the evaluation on CE 3D US and that on CE 3D CT achieved concordance in 61 lesions. Among them, 59 lesions were detected with the absence of tumor vessels and tumor enhancement and evaluated as adequate ablation, and the remaining two lesions were detected with residual tumors. The kappa value for agreement between the findings on the two modalities was 0.65. When 1-month CE 3D CT scans were used as reference standard, the sensitivity, specificity, and accuracy of 1-day CE 3D US for detecting adequate ablation were 97%, 100%, and 97%, respectively. Conclusion: By demonstrating the ablated areas and residual tumors in three dimensions, CE 3D US with Sonazoid was shown to be useful for evaluating the effect of RF ablation of HCCs, and there was good concordance with the results obtained by CE 3D CT. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Hepatocellular carcinoma (HCC), with the increasing diagnosis at an early stage as a result of the development of modern radiological imaging methods, is now one of the most commonly diagnosed carcinomas worldwide [1–3]. For some HCC patients with no possibility of resection or liver transplantation, imageguided radiofrequency (RF) ablation, has been demonstrated to be an effective, relatively safe, and technically feasible method of coagulating tumors by elevating the local temperature to above 60 ◦ C [4–6]. Completely ablated tumors become necrotic, however, because of certain factors, such as tumor location, imaging resolution, and operators’ experience, thermal ablation of the tumor may
∗ Corresponding author. Tel.: +81 45 261 5656; fax: +81 45 261 9492. E-mail address: [email protected] (K. Numata). 0720-048X/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2009.03.021
be incomplete, resulting in the persistence of residual tumor and the need for additional ablation session at a later time. Thus, accurate evaluation of the effects after RF ablation, whether the tumor has been ablated adequately, is critical for further treatment. Contrast-enhanced three-dimensional ultrasonography (CE 3D US) is a newly developed imaging technique that allows demonstration of the dynamic feature of HCCs and vessel structure in three orthogonal planes, and it has shown potential for characterizing hepatic tumors in recent studies [7–9]. However, to our knowledge, CE 3D US, which is expected to provide unique spatial views of ablated areas and residual tumors, has never been clarified in the evaluation of the effectiveness of RF ablation. A novel second-generation ultrasound contrast agent, Sonazoid (Daiichi Sankyo, Tokyo, Japan), which has been commercially available in Japan since January 2007, was used for CE 3D US imaging in our study. Sonazoid consists of microbubbles of perfluorobutane gas with phospholipid monolayer shells. The stable
92
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
Sonazoid microbubbles have been reported to be phagocytosed by reticuloendothelial cells in the liver parenchyma 5 min after administration, and thus they support a protracted contrast imaging [10–13]. The purpose of our study was to investigate the potential benefit of CE 3D US with a perfluorobutane-based contrast agent 1 day after ablation for evaluating the local effects of RF ablation of HCCs, in comparison with the findings of CE 3D computed tomography (CT) performed 1 month after RF ablation. 2. Materials and methods 2.1. Subjects This prospective study was performed with the approval of our institutional review board. Full informed consent was provided by all patients before the study. Between February 2007 and November 2007, 69 consecutive patients with HCCs less than 30 mm in diameter were admitted to our department for RF ablation treatment. Their HCCs were not eligible for surgery but were accessible for a percutaneous RF ablation and had not been treated previously. The final diagnosis of HCCs 2 cm or more in diameter was made on the basis of both typical CE CT and CE magnetic resonance imaging (MRI) findings, and the final diagnosis of those less than 2 cm in diameter was confirmed by percutaneous biopsy. Among these 69 patients, we included the subjects for the prospective study according to the following inclusion criteria: patients with a hypervascular tumor enhanced on dynamic radiological imaging; patients with a tumor located in a proper position for CE 3D US scanning free of interference caused by costal bones, abdominal gas or heart motion; patients with good compliableness for both CE 3D US and CE 3D CT examinations before and after ablation. After excluding three patients with HCCs in inappropriate locations for acquisition of satisfactory CE 3D US images, two patients who were allergic to the CT contrast agent, and one patient with a HCC that was hypovascular in three phases by both modalities, the remaining 63 patients were adopted as the subjects of this study. Multiple HCCs were detected in 23 subjects and a single tumor in the other 40 subjects. In the subjects with multiple HCCs, the largest HCC whose diagnosis had been confirmed and which was in a proper position for CE 3D US scanning was selected to evaluate before and after RFA ablation. The clinical characteristics of the subjects are shown in Table 1. Table 1 Clinical characteristics of the subjects with HCC lesions enrolled in this study. Characteristics No. of patients Age (mean, range, years)
n = 63 70, 53–80
Gender Male Female
n = 38 n = 25
Etiology of HCC Hepatitis C Hepatitis B Alcohol abuse
n = 56 n=4 n=3
Child-Pugh classification Class A Class B
n = 55 n=8
Diameter of lesions >2 cm ≤2 cm
n = 35 n = 28
Diameter of lesions (mean, range, mm)
22, 10–30
Diagnosis confirmed by Biopsy Radiological imaging
n = 29 n = 34
2.2. CE 3D US imaging Five to seven days before and 1 day after RF ablation, CE 3D US was performed by a sonographer with 10 years of experience in abdominal US. The LOGIQ 7 ultrasound imaging system (GE Healthcare, Milwaukee, WI) and a convex volume 4D3C-L probe with a 2.0–5.5-MHz frequency were used. With internal sectorial mechanical tilt movement, the probe held by the sonographer allowed automatic scanning of a volume of interest (VOI). The position and size of the VOI could be adjusted before scanning so that it would cover the desired region. The LOGIQ 7 ultrasound imaging system is equipped with Autosweep 3D and Static 3D functionalities, and they were used for image acquisition by CE 3D US. Before the CE 3D US scanning, all subjects received an intravenous bolus injection of 0.2 mL of Sonazoid, followed by 2 mL of 5% glucose solution, and subsequent infusion of 5% glucose solution at 10 mL/min. The contrast harmonic angio (CHA) mode (mechanical index = 0.5–0.9) at 8–13 frames per second was used as the insonation technique for CE 3D US. When the CE 3D US was performed, the reference images for the location of the lesions were shown on the same screen by the harmonic ultrasonography. The CE 3D US images were acquired during three contrast phases, consisting of an early phase (10–60 s after contrast medium injection), a middle phase (80–120 s after contrast medium injection), and a late phase (more than 5 min after contrast medium injection). The data acquired were stored as cineloops in the hard disk of the ultrasound imaging system. After CE 3D scanning, the 3D images were reconstructed using the functionalities of the ultrasound imaging system. In each contrast phase tomographic ultrasound images (TUI) in view of parallel slices were reconstructed in three orthogonal planes, i.e., plane A, which could be translated from front to back in the VOI, plane B, which could be translated from right to left, and plane C, which could be translated from up to down. The distance between two adjacent slices could be adjusted in order to show the desired regions. Sonographic angiograms were reconstructed in angio-like views during the early phase and middle phase by using various rendering modes. The maximum intensity mode for displaying the maximum intensity grey value of the VOI, mixed with the surface mode for displaying the grey value on the surface of the object, was used to visualize tumor vessels and early tumor enhancement before RF ablation and to detect residual viable portions of hypervascular tumors after RF ablation, while the average intensity mode for displaying the average intensity grey value of the VOI, mixed with the surface mode, was employed to describe the unenhanced areas, such as the coagulated areas with perfusion defect after treatment. TUI in three orthogonal planes and sonographic angiogram images with raw volume data were stored in the hard disk of the ultrasound imaging system. 2.3. CE 3D CT imaging Five to seven days before the RF ablation, 1 month after the RF ablation, and during the follow-up period CT scanning was performed with a commercially available CT scanner (16-channel multi-detector-row CT scanner; Toshiba Medical Systems Co., Ltd., Tokyo, Japan) with the following protocol: tube voltage, 120 kV; tube current, auto mA exposure setting; reconstruction section and interval thickness, 5 mm; detector configuration, 16 mm × 1 mm; pitch, 15; and 0.5 s per rotation. The subjects were divided into a group weighing under 70 kg, who were injected with a 300-mgI/mL dose of the nonionic contrast medium iopamidol (Iopamiron 300; Bayer Healthcare, Osaka, Japan) and a group weighing 70 kg or more, who were injected with a 370-mgI/mL dose of iopamidol. A catheter placed in the peripheral vein of the antecubital fossa was used to administer 100 mL of contrast medium at a rate of
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
3 mL/s with a power injector (Dual shot GX, Nemoto Kyorindo, Tokyo, Japan). The trigger point for starting the arterial phase scan was set at a 230-Hounsfield unit enhancement over the baseline attenuation of the abdominal aorta, and was confirmed with an automatic-bolus-tracking program (RealPrep; Toshiba Medical Systems Co., Ltd., Tokyo, Japan). Scanning for the portal venous phase and the equilibrium phase was performed at 70 s and 180 s, respectively, after the start of contrast agent injection. The data obtained by CT scanning were transferred to a Zio M900 workstation (Zio software, Tokyo, Japan), and a technician blinded to the final diagnosis, clinical information, and other radiological findings reconstructed both multiplanar images and CT angiograms. In each contrast phase multiplanar images were created in three orthogonal planes, i.e., the axial plane, sagittal, and coronal planes, and CT angiograms were reconstructed in the arterial phase and portal phase by the maximum-intensity-projection technique. All tomographic images and CT angiograms were stored in the Zio M900 workstation. 2.4. RF ablation procedure RF ablation was performed under real-time US guidance using LOGIQ 7 ultrasound system and a 3.5-MHz convex probe (GE Healthcare, Milwaukee, WI), and by one physician who had 10 years of experience in RF ablation of HCCs. To ablate 41 of the tumors, a 20-cm-long 17-gauge cool-tip radiofrequency electrode with a 2- or 3-cm-long exposed metallic tip (Cool-tip Needle; Radionics, Burlington, MA, USA) was promptly inserted into the targeted tumor, and the RF ablation was performed with an RF generator system (Radionics, Burlington, MA). The other 22 tumors were ablated with hooked, 15-gauge, 25-cm-long electrodes (LeVeen Needle Electrode; Radiotherapeutics, Mountain View, CA, USA) and a generator (RTC 2000; Boston Scientific Japan, Tokyo, Japan). The subjects held their breath for a few seconds during the RF ablation.
93
2.5. Image analysis The strategy of treatment and imaging evaluation is shown in Fig. 1. Five to seven days prior to RF ablation, CE 3D US and CE 3D CT were performed to identify tumor location and tumor enhancement. One day after the RF ablation, CE 3D US was performed to evaluate the therapeutic effects. If any residual tumor was detected, additional RF ablation procedures were performed during the next 2–3 days to ensure adequate ablation of the tumor. One month after RF ablation (the final procedure for tumors that required more than one RF ablation), all of the lesions were scanned by CE 3D CT. If no residual tumor was detected on the 1-month CE 3D CT images, follow-up CE 3D CT was performed every 3 months after the RF ablation. The follow-up period of this study was ended in October 2008. To evaluate the effect of treatment, the 3D images acquired by each modality before and after RF ablation were read by two experienced gastrointestinal radiologists, neither of whom was involved in the RF ablation procedure and both of whom were blinded to clinical information and other radiological findings. The images were reviewed independently: the CE 3D US images on the ultrasound imaging system, and CE 3D CT images on the Zio M900 workstation. As they reviewed the 3D images, the two radiologists had the option of interactively translating the multiplanar images and changing the angle of view of the angiograms, if necessary. Finally, the two radiologists conferred and arrived at a consensus. When the lesion sites were evaluated on the CE 3D US images 1 day after the RF ablation, the ablation was evaluated as adequate if after ablation a non-enhancing area seen in the early phase, middle phase, and late phase covered the hypervascular enhancement seen in the early phase and middle phase before ablation. Residual tumor on CE 3D US images was diagnosed when an area within the tumor was detected with hypervascular enhancement in the
Fig. 1. Flow chart of treatment and imaging strategy. Pre-treatment and post-treatment imaging was performed by contrast-enhanced three-dimensional ultrasonography (CE 3D US) and CE 3D computed tomography (CT).
94
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
Fig. 2. A 70-year-old woman with hepatocellular carcinoma in the inferior posterior segment of the right liver lobe. (a–h) Contrast-enhanced three-dimensional ultrasonography images show that before radiofrequency (RF) ablation the HCC located adjacent to the right posterior portal vein was distinctly enhanced in the early phase, as shown on tomographic ultrasound images in plane A with a slice distance of 1.0 mm (a), plane B with a slice distance of 1.5 mm (b), and plane C with a slice distance of 1.0 mm (c) and on the sonographic angiogram (d) rendered by using maximum intensity mixed with surface mode. One day after treatment, adequate ablation in the absence of enhancement was detected as shown in the middle phase on the tomographic ultrasound images in plane A with a slice distance of 1.9 mm (e), plane B with a slice distance of 2.3 mm (f), and plane C with a slice distance of 2.4 mm (g), and on the sonographic angiogram (h) rendered by using average intensity mixed with surface mode. (i–p) Contrast-enhanced three-dimensional computed tomography images show that in the arterial phase before RF ablation the lesion was detected as an area of high attenuation on multiplanar images in the transverse plane (i), coronal plane (j) and sagittal plane (k), and on the CT angiogram (l) reconstructed by using the maximum-intensity-projection. One month after treatment adequate ablation was detected as an area of low attenuation in the arterial phase on multiplanar images in the transverse plane (m), coronal plane (n), and sagittal plane (o), and on the CT angiogram (p) reconstructed by using maximum-intensity-projection method. During a 20-month follow-up period, there was no local tumor progression detected on CE 3D CT. Arrowheads indicate the location of the original tumor before treatment or the ablated area after treatment.
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
95
Table 2 Evaluation by contrast-enhanced three-dimensional US and contrast-enhanced three-dimensional CT after radiofrequency ablation of HCCs. No. of lesions
One-day CE 3D US
One-month CE 3D CT
Follow-up CE 3D CT
n = 58 n=1 n=1 n=2 n=1
Adequate ablation Residual tumora Adequate ablation Residual tumora Residual tumora
Adequate ablation Adequate ablation Adequate ablation Residual tumorc Adequate ablation
No local tumor progression during follow-up period (range 11–21 months) Local tumor progression on 3-month CE 3D CTb Local tumor progression on 6-month CE 3D CTb No local tumor progression during 15-month follow-up
Note: “One-day CE 3D US” means contrast-enhanced three-dimensional ultrasonography performed 1 day after radiofrequency ablation (or the final ablation if multiple procedures were performed on the same lesion). “One-month CE 3D CT” and “follow-up CE 3D CT” mean contrast-enhanced three-dimensional computed tomography performed 1 month after radiofrequency ablation and that performed every 3 months after the RF ablation, respectively. a The lesion with residual tumors detected on 1-day CE 3D US, and the subject did not give their consent for the additional radiofrequency ablation. b The lesion with local tumor progression on the follow-up CE 3D CT was treated with the radiofrequency ablation or transarterial chemoembolization. c The lesion with residual tumors detected on 1-month CE 3D CT was then treated with the radiofrequency ablation.
early phase and middle phase, and was hypoechoic or isoechoic in the late phase of CE 3D US after ablation. When the lesion sites were evaluated on CE 3D CT after RF ablation, the images were evaluated as showing adequate ablation if after ablation a lowattenuation area was observed in the arterial phase, portal venous phase, and equilibrium phase covering the high-attenuation area seen in the arterial phase before ablation. Residual tumor on 1month CE 3D CT images was diagnosed if an area within the tumor showed high attenuation in the arterial phase and low-attenuation area in the portal venous phase, equilibrium phase, or both after ablation. Local tumor progression was evaluated if enhancement in the margin of the ablated areas in the arterial phase showed low attenuation in the portal venous phase, equilibrium phase, or both on the CE 3D CT images acquired every 3 months as follow-up examinations. 2.6. Statistical analysis Taken consideration of the findings on 1-month CE 3D CT images as a reference standard, the sensitivity, specificity, and accuracy of CE 3D US for detecting the adequate ablation was calculated. The concordance between the two modalities for detecting residual tumors was evaluated by the kappa test. Agreement was graded according to the kappa values in parentheses as poor (<0.20), moderate (0.20–0.40), fair (0.40–0.60), good (0.60–0.80), or excellent (0.80–1.0). The SPSS version 11.0 software package (SPSS, Tokyo, Japan) was used. 3. Results All 63 HCCs were treated by RF ablation and evaluated by using CE 3D US and CE 3D CT examinations before and after treatment. No subjects were excluded from the analysis during the follow-up period (range 3–21 months, mean, 15 months). Of the 63 HCCs, 54 were treated by a single RF ablation, and the other 9 HCCs were treated by 2 or 3 ablation procedures each to ensure adequate ablation. On the CE 3D US images before RF ablation all 63 HCCs appeared as hypervascular areas with diffused tumor enhancement in the early or middle phase (Fig. 2a–d) and as hypoechoic (n = 60) or isoechoic (n = 3) areas in the late phase, and on the CE 3D CT images they appeared as high-attenuation areas in the arterial phase (Fig. 2i–l) and as low-attenuation areas in the portal or equilibrium phase. Inter-reader variation for evaluating the effects of RF ablation was minimal. In 58 of the 63 lesions the two readers’ evaluation were the same. Different evaluations between two readers achieved in consensus in the other five lesions, including four evaluated as adequate ablation and one as residual tumors. The evaluation on CE 3D US and that on 1-month CE 3D CT achieved concordance in 61 lesions. As shown in Table 2, 59 of the 61 lesions showed no evidence of residual tumor in the ablated
areas on the CE 3D US images 1 day after the RF ablation or on the CE 3D CT images 1 month after RF ablation, and adequate ablation of 58 of the 59 lesions was confirmed by the subsequent follow-up CE 3D CT examinations. The absence of tumor vessels and tumor enhancement in the early phase, middle phase and late phase of the 1-day CE 3D US demonstrated the adequate ablation of HCCs (Fig. 2e–h), which was manifested as a low-attenuation area in the arterial phase, portal venous phase, and equilibrium phase of the CE 3D CT (Fig. 2m–p). In the remaining 2 of the 61 lesions, residual tumor due to inadequate ablation was detected on both 1-day CE 3D US and 1-month CE 3D CT. On 1-day CE 3D US residual tumor showed the hypervascular enhancement in the early phase and middle phase and hypoechoic in the late phase, however, additional RF ablation was not performed because the subjects did not give their consent. After 1 month, in both cases, residual tumor of high attenuation in the arterial phase and low attenuation in both the portal venous phase and equilibrium phases was detected in the peripheral ablated areas on the CE 3D CT. There was discordance between the evaluations by the two modalities in 2 of the 63 lesions. Residual tumor in both lesions was detected by CE 3D US 1 day after the RF ablation, and additional RF ablation was not performed because the subjects did not give their consent. The 1-month CE 3D CT showed no evidence of residual tumor in the ablated areas. Local tumor progression on one of these two lesions was identified during the follow-up period of CE 3D CT, but there was no evidence of tumor progression on the other lesion (Table 2). The kappa value for agreement between the two modalities was 0.65. Comparison of the results of 1-day CE 3D US with those of the 1-month CE 3D CT showed that the sensitivity of CE 3D US for detecting adequate ablation was 97% (59/61), specificity100% (2/2), and the accuracy 97% (61/63).
4. Discussion Advances in imaging techniques and contrast agents have improved the role of radiological imaging in evaluating the effect of RF ablation of HCCs [14–17]. On the dynamic radiological images, the necrotic areas after RF ablation of HCCs appeared as an absence of contrast agent perfusion, while residual portion of hypervascular tumors were usually enhanced in the vascular contrast phase and the contrast agent was washed out in the following parenchyma contrast phase [18–20]. CE CT is accepted as the gold standard for assessing whether HCC tumors are coagulated adequately, and it is performed routinely to evaluate the effect of ablation and for the long-term follow-up [21,22]. However, immediately after the procedure of RF ablation and during the early follow-up time, on CE CT images, the presence of hyperattenuation due to the dehydration produced by coagulative necrosis, and periablational enhancement due to reactive hyperemia, arteriovenous shunts, or fibrosis/giant
96
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
cell reaction, hinder accurate use of this modality to evaluate RF ablation [23–25]. Compared to CE CT images, CE US images acquired in the early period after RF ablation are accompanied with few artifacts and have been proven to be useful for the early evaluation of RF ablation [26–28]. Wen et al. found that CE 2D US performed 5–7 days after RF ablation of 107 HCCs with the CHA mode and contrast agent Levovist had a sensitivity, specificity and accuracy of 95.3%, 100% and 98.1%, respectively, for detecting residual tumor [29]. In the present study, we used a newly developed imaging technique, CE 3D US, which has been reported to facilitate convenient and accurate observation of tortuous tumor vessels and tumor enhancement in a 3D view [9], to demonstrate residual tumor and thereby determine whether the tumor had been ablated completely or only partially. The reconstructed multiplanar TUI in the three orthogonal planes allowed the observers to detect different portions of tumors by synchronized presentation of multiple layers, while the sonographic angiograms spatially demonstrated the structures in the entire VOI with various view-angles. The 3D US images allowed the observers to compare the HCC tumors before treatment with the ablated areas after treatment in a 3D visualization. When the ablated areas and residual tumor were detected accurately in the three orthogonal planes of multiplanar images, sonographic angiograms could be used as a supplementary tool to view the lesion and surrounding anatomical structures in an angio-like perspective. CE 3D CT images were reported to be useful for preoperative visualization of the anatomic structures and for the intraoperative imaging guidance [30,31]. In our study CE 3D CT performed 1 month after RF ablation were regarded as the reference standard for the evaluation of RF ablation. Comparison with the results on 1-month CE 3D CT showed that CE 3D US had high sensitivity, specificity, and accuracy for detecting adequate ablation, thereby revealing a potential role for CE 3D US in assessing the effect of RF ablation. Unlike CT contrast media, the contrast agent Sonazoid used in our study does not disperse into the interstitial spaces. And with the assistance of CHA mode and high MI contrast conditions, which reduce microbubbles in microvessels but not in relatively large vessels, such as tumor vessels and portal veins [10], Sonazoid-enhanced 3D US facilitated elaborate observation of tumor vessels. Of the four lesions in which residual tumor was detected by 1-day CE 3D US, two were confirmed by 1-month CE 3D CT and then treated by additional RF ablation or transarterial chemoembolization, one was evaluated as adequate ablation by 1-month CE 3D CT and detected with local tumor progression in the marginal ablated areas by CE 3D CT after 3 months and then treated by additional RF ablation, and the remaining one was detected without local enhancement on 1month and the follow-up CE 3D CT. Our results to some extent imply CE 3D US is a sensitive tool for detecting residual tumor in the early time after RF ablation, although there may be some false-positive findings. Our study had some limitations. First, some lesions with residual tumor by 1-day CE 3D US after the initial treatment were subjected to the additional RF ablation to achieve adequate coagulation, and thus only four lesions with residual tumors after ablation were evaluated on CE 3D US, because these subjects did not consent to additional RF ablation. The small number of positive findings by CE 3D US and the relatively small study population may have affected the accuracy of assessment by CE 3D US. Second, one hypovascular HCC that was not enhanced on either imaging modality was excluded because it is difficult to evaluate the effect of RF ablation of hypovascular lesions on the basis of assessing tumor enhancement alone. Third, our initial study emphasized the local effect of RF ablation, and long-term follow-up to detect distant newly developed HCCs by CE 3D US needs further investigation.
In conclusion, because of its good concordance with the results of CE 3D CT, CE 3D US, which is convenient to perform and does not involve radiation exposure, can be used to evaluate the effects of RF ablation of HCCs with unique 3D view. References [1] Parkin DM, Bray F, Ferlay J, Pisani P. Estimating the world cancer burden: globocan 2000. Int J Cancer 2001;94:153–6. [2] Lau WY, Lai EC. Hepatocellular carcinoma: current management and recent advances. Hepatobiliary Pancreat Dis Int 2008;7:237–57. [3] El-Serag HB, Marrero JA, Rudolph L, Reddy KR. Diagnosis and treatment of hepatocellular carcinoma. Gastroenterology 2008;134:1752–63. [4] Veltri A, Moretto P, Doriguzzi A, Pagano E, Carrara G, Gandini G. Radiofrequency thermal ablation (RFA) after transarterial chemoembolization (TACE) as a combined therapy for unresectable non-early hepatocellular carcinoma (HCC). Eur Radiol 2006;16:661–9. [5] Howard JH, Tzeng CW, Smith JK, et al. Radiofrequency ablation for unresectable tumors of the liver. Am Surg 2008;74:594–600. [6] Crocetti L, Lencioni R. Thermal ablation of hepatocellular carcinoma. Cancer Imaging 2008;8:19–26. [7] Dietrich CF. 3D real time contrast enhanced ultrasonography—a new technique. Rofo 2002;174:160–3. [8] Ohto M, Kato H, Tsujii H, Maruyama H, Matsutani S, Yamagata H. Vascular flow patterns of hepatic tumors in contrast-enhanced 3-dimensional fusion ultrasonography using plane shift and opacity control modes. J Ultrasound Med 2005;24:49–57. [9] Luo W, Numata K, Morimoto M, et al. Clinical utility of contrast-enhanced three-dimensional ultrasound imaging with Sonazoid: findings on hepatocellular carcinoma lesions. Eur J Radiol. doi:10.1016/j.ejrad.2008.09.003 [accessed on October 17, 2008]. [10] Numata K, Morimoto M, Ogura T, et al. Ablation therapy guided by contrastenhanced sonography with Sonazoid for hepatocellular carcinoma lesions not detected by conventional sonography. J Ultrasound Med 2008;27:395–406. [11] Hatanaka K, Kudo M, Minami Y, et al. Differential diagnosis of hepatic tumors: value of contrast-enhanced harmonic sonography using the newly developed contrast agent, Sonazoid. Intervirology 2008;51 S1:61–9. [12] Nakano H, Ishida Y, Hatakeyama T, et al. Contrast-enhanced intraoperative ultrasonography equipped with late Kupffer-phase image obtained by sonazoid in patients with colorectal liver metastases. World J Gastroenterol 2008;14:3207–11. [13] Edey AJ, Ryan SM, Beese RC, Gordon P, Sidhu PS. Ultrasound imaging of liver metastases in the delayed parenchymal phase following administration of Sonazoid using a destructive mode technique (Agent Detection Imaging). Clin Radiol 2008;63:1112–20. [14] Buscarini E, Savoia A, Brambilla G, et al. Radiofrequency thermal ablation of liver tumors. Eur Radiol 2005;15:884–94. [15] Schima W, Ba-Ssalamah A, Kurtaran A, Schindl M, Gruenberger T. Posttreatment imaging of liver tumours. Cancer Imaging 2007;7:28–36. [16] 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. [17] McGahan JP, Khatri VP. Imaging findings after liver resection by using radiofrequency parenchymal coagulation devices: initial experiences. Radiology 2008;247:896–902. [18] Ding H, Kudo M, Onda H, et al. Evaluation of posttreatment response of hepatocellular carcinoma with contrast-enhanced coded phase-inversion harmonic US: comparison with dynamic CT. Radiology 2001;221:721–30. [19] Tsuda M, Majima K, Yamada T. Hepatocellular carcinoma after radiofrequency ablation therapy: dynamic CT evaluation of treatment. Clin Imaging 2001;25:409–15. [20] Fujioka C, Horiguchi J, Ishifuro M, et al. A feasibility study: evaluation of radiofrequency ablation therapy to hepatocellular carcinoma using image registration of preoperative and postoperative CT. Acad Radiol 2006;13:986–94. [21] Lim HS, Jeong YY, Kang HK, Kim JK, Park JG. Imaging features of hepatocellular carcinoma after transcatheter arterial chemoembolization and radiofrequency ablation. AJR Am J Roentgenol 2006;187:W341–9. [22] Geyik S, Akhan O, Abbaso˘glu O, et al. Radiofrequency ablation of unresectable hepatic tumors. Diagn Interv Radiol 2006;12:195–200. [23] Goldberg SN, Gazelle GS, Mueller PR. Thermal ablation therapy for focal malignancy: a unified approach to underlying principles, techniques, and diagnostic imaging guidance. AJR Am J Roentgenol 2000;174:323–31. [24] Dill-Macky MJ, Asch M, Burns P, Wilson S. Radiofrequency ablation of hepatocellular carcinoma: predicting success using contrast-enhanced sonography. AJR Am J Roentgenol 2006;186:S287–95. [25] Filippone A, Iezzi R, Di Fabio F, Cianci R, Grassedonio E, Storto ML. Multidetectorrow computed tomography of focal liver lesions treated by radiofrequency ablation: spectrum of findings at long-term follow-up. J Comput Assist Tomogr 2007;31:42–52. [26] Choi D, Lim HK, Lee WJ, et al. Radiofrequency ablation of liver cancer: early evaluation of therapeutic response with contrast-enhanced ultrasonography. Korean J Radiol 2004;5:185–98. [27] Kim CK, Choi D, Lim HK, et al. Therapeutic response assessment of percutaneous radiofrequency ablation for hepatocellular carcinoma: util-
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97 ity of contrast-enhanced agent detection imaging. Eur J Radiol 2005;56: 66–73. [28] Vilana R, Bianchi L, Varela M, et al. Is microbubble-enhanced ultrasonography sufficient for assessment of response to percutaneous treatment in patients with early hepatocellular carcinoma? Eur Radiol 2006;16:2454–62. [29] Wen YL, Kudo M, Zheng RQ, et al. Radiofrequency ablation of hepatocellular carcinoma: therapeutic response using contrast-enhanced coded phase-inversion harmonic sonography. AJR Am J Roentgenol 2003;181:57–63.
97
[30] Endo I, Shimada H, Sugita M, et al. Role of three-dimensional imaging in operative planning for hilar cholangiocarcinoma. Surgery 2007;142: 666–75. [31] Dong Q, Xu W, Jiang B, et al. Clinical applications of computerized tomography 3-D reconstruction imaging for diagnosis and surgery in children with large liver tumors or tumors at the hepatic hilum. Pediatr Surg Int 2007;23: 1045–50.
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Role of Sonazoid-enhanced three-dimensional ultrasonography in the evaluation of percutaneous radiofrequency ablation of hepatocellular carcinoma Wen Luo a,b , Kazushi Numata a,∗ , Manabu Morimoto a , Takashi Oshima a , Michio Ueda a , Masahiro Okada c , Shigeo Takebayashi d , Xiaodong Zhou b , Katsuaki Tanaka a a
Gastroenterological Center, Yokohama City University Medical Center, 4-57 Urafune-cho, Minami-ku, Yokohama, Kanagawa, 232-0024, Japan Department of Ultrasound, Xijing Hospital, Fourth Military Medical University, 15th Changle Xi Road, Xi’an, ShaanXi, 710032, China c Department of Radiology, Kinki University School of Medicine, 377-2 Ohno-Higashi, Osaka-Sayama, Osaka, 589-8511, Japan d Department of Radiology, Yokohama City University Medical Center, 4-57 Urafune-cho, Minami-ku, Yokohama, Kanagawa, 232-0024, Japan b
a r t i c l e
i n f o
Article history: Received 23 January 2009 Accepted 13 March 2009 Keywords: Three-dimensional ultrasonography Three-dimensional computed tomography Contrast agent Radiofrequency ablation Hepatocellular carcinoma
a b s t r a c t Objective: We investigated contrast-enhanced three-dimensional ultrasonography (CE 3D US) with contrast agent Sonazoid for evaluating the effect of percutaneous radiofrequency (RF) ablation of hepatocellular carcinomas (HCCs). Methods: 63 HCCs were treated by US-guided percutaneous RF ablation. CE 3D US after bolus injection of 0.2 mL of Sonazoid was performed 5–7 days before and 1 day after RF ablation. CE 3D computed tomography (CT) was performed 5–7 days before and 1 month after the ablation, and during the followup period. Multiplanar images in three orthogonal planes and US/CT angiograms were reconstructed on both modalities. Two blinded observers reviewed the images on both modalities to evaluate the ablation effects. Results: After RF ablation, the evaluation on CE 3D US and that on CE 3D CT achieved concordance in 61 lesions. Among them, 59 lesions were detected with the absence of tumor vessels and tumor enhancement and evaluated as adequate ablation, and the remaining two lesions were detected with residual tumors. The kappa value for agreement between the findings on the two modalities was 0.65. When 1-month CE 3D CT scans were used as reference standard, the sensitivity, specificity, and accuracy of 1-day CE 3D US for detecting adequate ablation were 97%, 100%, and 97%, respectively. Conclusion: By demonstrating the ablated areas and residual tumors in three dimensions, CE 3D US with Sonazoid was shown to be useful for evaluating the effect of RF ablation of HCCs, and there was good concordance with the results obtained by CE 3D CT. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Hepatocellular carcinoma (HCC), with the increasing diagnosis at an early stage as a result of the development of modern radiological imaging methods, is now one of the most commonly diagnosed carcinomas worldwide [1–3]. For some HCC patients with no possibility of resection or liver transplantation, imageguided radiofrequency (RF) ablation, has been demonstrated to be an effective, relatively safe, and technically feasible method of coagulating tumors by elevating the local temperature to above 60 ◦ C [4–6]. Completely ablated tumors become necrotic, however, because of certain factors, such as tumor location, imaging resolution, and operators’ experience, thermal ablation of the tumor may
∗ Corresponding author. Tel.: +81 45 261 5656; fax: +81 45 261 9492. E-mail address: [email protected] (K. Numata). 0720-048X/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2009.03.021
be incomplete, resulting in the persistence of residual tumor and the need for additional ablation session at a later time. Thus, accurate evaluation of the effects after RF ablation, whether the tumor has been ablated adequately, is critical for further treatment. Contrast-enhanced three-dimensional ultrasonography (CE 3D US) is a newly developed imaging technique that allows demonstration of the dynamic feature of HCCs and vessel structure in three orthogonal planes, and it has shown potential for characterizing hepatic tumors in recent studies [7–9]. However, to our knowledge, CE 3D US, which is expected to provide unique spatial views of ablated areas and residual tumors, has never been clarified in the evaluation of the effectiveness of RF ablation. A novel second-generation ultrasound contrast agent, Sonazoid (Daiichi Sankyo, Tokyo, Japan), which has been commercially available in Japan since January 2007, was used for CE 3D US imaging in our study. Sonazoid consists of microbubbles of perfluorobutane gas with phospholipid monolayer shells. The stable
92
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
Sonazoid microbubbles have been reported to be phagocytosed by reticuloendothelial cells in the liver parenchyma 5 min after administration, and thus they support a protracted contrast imaging [10–13]. The purpose of our study was to investigate the potential benefit of CE 3D US with a perfluorobutane-based contrast agent 1 day after ablation for evaluating the local effects of RF ablation of HCCs, in comparison with the findings of CE 3D computed tomography (CT) performed 1 month after RF ablation. 2. Materials and methods 2.1. Subjects This prospective study was performed with the approval of our institutional review board. Full informed consent was provided by all patients before the study. Between February 2007 and November 2007, 69 consecutive patients with HCCs less than 30 mm in diameter were admitted to our department for RF ablation treatment. Their HCCs were not eligible for surgery but were accessible for a percutaneous RF ablation and had not been treated previously. The final diagnosis of HCCs 2 cm or more in diameter was made on the basis of both typical CE CT and CE magnetic resonance imaging (MRI) findings, and the final diagnosis of those less than 2 cm in diameter was confirmed by percutaneous biopsy. Among these 69 patients, we included the subjects for the prospective study according to the following inclusion criteria: patients with a hypervascular tumor enhanced on dynamic radiological imaging; patients with a tumor located in a proper position for CE 3D US scanning free of interference caused by costal bones, abdominal gas or heart motion; patients with good compliableness for both CE 3D US and CE 3D CT examinations before and after ablation. After excluding three patients with HCCs in inappropriate locations for acquisition of satisfactory CE 3D US images, two patients who were allergic to the CT contrast agent, and one patient with a HCC that was hypovascular in three phases by both modalities, the remaining 63 patients were adopted as the subjects of this study. Multiple HCCs were detected in 23 subjects and a single tumor in the other 40 subjects. In the subjects with multiple HCCs, the largest HCC whose diagnosis had been confirmed and which was in a proper position for CE 3D US scanning was selected to evaluate before and after RFA ablation. The clinical characteristics of the subjects are shown in Table 1. Table 1 Clinical characteristics of the subjects with HCC lesions enrolled in this study. Characteristics No. of patients Age (mean, range, years)
n = 63 70, 53–80
Gender Male Female
n = 38 n = 25
Etiology of HCC Hepatitis C Hepatitis B Alcohol abuse
n = 56 n=4 n=3
Child-Pugh classification Class A Class B
n = 55 n=8
Diameter of lesions >2 cm ≤2 cm
n = 35 n = 28
Diameter of lesions (mean, range, mm)
22, 10–30
Diagnosis confirmed by Biopsy Radiological imaging
n = 29 n = 34
2.2. CE 3D US imaging Five to seven days before and 1 day after RF ablation, CE 3D US was performed by a sonographer with 10 years of experience in abdominal US. The LOGIQ 7 ultrasound imaging system (GE Healthcare, Milwaukee, WI) and a convex volume 4D3C-L probe with a 2.0–5.5-MHz frequency were used. With internal sectorial mechanical tilt movement, the probe held by the sonographer allowed automatic scanning of a volume of interest (VOI). The position and size of the VOI could be adjusted before scanning so that it would cover the desired region. The LOGIQ 7 ultrasound imaging system is equipped with Autosweep 3D and Static 3D functionalities, and they were used for image acquisition by CE 3D US. Before the CE 3D US scanning, all subjects received an intravenous bolus injection of 0.2 mL of Sonazoid, followed by 2 mL of 5% glucose solution, and subsequent infusion of 5% glucose solution at 10 mL/min. The contrast harmonic angio (CHA) mode (mechanical index = 0.5–0.9) at 8–13 frames per second was used as the insonation technique for CE 3D US. When the CE 3D US was performed, the reference images for the location of the lesions were shown on the same screen by the harmonic ultrasonography. The CE 3D US images were acquired during three contrast phases, consisting of an early phase (10–60 s after contrast medium injection), a middle phase (80–120 s after contrast medium injection), and a late phase (more than 5 min after contrast medium injection). The data acquired were stored as cineloops in the hard disk of the ultrasound imaging system. After CE 3D scanning, the 3D images were reconstructed using the functionalities of the ultrasound imaging system. In each contrast phase tomographic ultrasound images (TUI) in view of parallel slices were reconstructed in three orthogonal planes, i.e., plane A, which could be translated from front to back in the VOI, plane B, which could be translated from right to left, and plane C, which could be translated from up to down. The distance between two adjacent slices could be adjusted in order to show the desired regions. Sonographic angiograms were reconstructed in angio-like views during the early phase and middle phase by using various rendering modes. The maximum intensity mode for displaying the maximum intensity grey value of the VOI, mixed with the surface mode for displaying the grey value on the surface of the object, was used to visualize tumor vessels and early tumor enhancement before RF ablation and to detect residual viable portions of hypervascular tumors after RF ablation, while the average intensity mode for displaying the average intensity grey value of the VOI, mixed with the surface mode, was employed to describe the unenhanced areas, such as the coagulated areas with perfusion defect after treatment. TUI in three orthogonal planes and sonographic angiogram images with raw volume data were stored in the hard disk of the ultrasound imaging system. 2.3. CE 3D CT imaging Five to seven days before the RF ablation, 1 month after the RF ablation, and during the follow-up period CT scanning was performed with a commercially available CT scanner (16-channel multi-detector-row CT scanner; Toshiba Medical Systems Co., Ltd., Tokyo, Japan) with the following protocol: tube voltage, 120 kV; tube current, auto mA exposure setting; reconstruction section and interval thickness, 5 mm; detector configuration, 16 mm × 1 mm; pitch, 15; and 0.5 s per rotation. The subjects were divided into a group weighing under 70 kg, who were injected with a 300-mgI/mL dose of the nonionic contrast medium iopamidol (Iopamiron 300; Bayer Healthcare, Osaka, Japan) and a group weighing 70 kg or more, who were injected with a 370-mgI/mL dose of iopamidol. A catheter placed in the peripheral vein of the antecubital fossa was used to administer 100 mL of contrast medium at a rate of
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
3 mL/s with a power injector (Dual shot GX, Nemoto Kyorindo, Tokyo, Japan). The trigger point for starting the arterial phase scan was set at a 230-Hounsfield unit enhancement over the baseline attenuation of the abdominal aorta, and was confirmed with an automatic-bolus-tracking program (RealPrep; Toshiba Medical Systems Co., Ltd., Tokyo, Japan). Scanning for the portal venous phase and the equilibrium phase was performed at 70 s and 180 s, respectively, after the start of contrast agent injection. The data obtained by CT scanning were transferred to a Zio M900 workstation (Zio software, Tokyo, Japan), and a technician blinded to the final diagnosis, clinical information, and other radiological findings reconstructed both multiplanar images and CT angiograms. In each contrast phase multiplanar images were created in three orthogonal planes, i.e., the axial plane, sagittal, and coronal planes, and CT angiograms were reconstructed in the arterial phase and portal phase by the maximum-intensity-projection technique. All tomographic images and CT angiograms were stored in the Zio M900 workstation. 2.4. RF ablation procedure RF ablation was performed under real-time US guidance using LOGIQ 7 ultrasound system and a 3.5-MHz convex probe (GE Healthcare, Milwaukee, WI), and by one physician who had 10 years of experience in RF ablation of HCCs. To ablate 41 of the tumors, a 20-cm-long 17-gauge cool-tip radiofrequency electrode with a 2- or 3-cm-long exposed metallic tip (Cool-tip Needle; Radionics, Burlington, MA, USA) was promptly inserted into the targeted tumor, and the RF ablation was performed with an RF generator system (Radionics, Burlington, MA). The other 22 tumors were ablated with hooked, 15-gauge, 25-cm-long electrodes (LeVeen Needle Electrode; Radiotherapeutics, Mountain View, CA, USA) and a generator (RTC 2000; Boston Scientific Japan, Tokyo, Japan). The subjects held their breath for a few seconds during the RF ablation.
93
2.5. Image analysis The strategy of treatment and imaging evaluation is shown in Fig. 1. Five to seven days prior to RF ablation, CE 3D US and CE 3D CT were performed to identify tumor location and tumor enhancement. One day after the RF ablation, CE 3D US was performed to evaluate the therapeutic effects. If any residual tumor was detected, additional RF ablation procedures were performed during the next 2–3 days to ensure adequate ablation of the tumor. One month after RF ablation (the final procedure for tumors that required more than one RF ablation), all of the lesions were scanned by CE 3D CT. If no residual tumor was detected on the 1-month CE 3D CT images, follow-up CE 3D CT was performed every 3 months after the RF ablation. The follow-up period of this study was ended in October 2008. To evaluate the effect of treatment, the 3D images acquired by each modality before and after RF ablation were read by two experienced gastrointestinal radiologists, neither of whom was involved in the RF ablation procedure and both of whom were blinded to clinical information and other radiological findings. The images were reviewed independently: the CE 3D US images on the ultrasound imaging system, and CE 3D CT images on the Zio M900 workstation. As they reviewed the 3D images, the two radiologists had the option of interactively translating the multiplanar images and changing the angle of view of the angiograms, if necessary. Finally, the two radiologists conferred and arrived at a consensus. When the lesion sites were evaluated on the CE 3D US images 1 day after the RF ablation, the ablation was evaluated as adequate if after ablation a non-enhancing area seen in the early phase, middle phase, and late phase covered the hypervascular enhancement seen in the early phase and middle phase before ablation. Residual tumor on CE 3D US images was diagnosed when an area within the tumor was detected with hypervascular enhancement in the
Fig. 1. Flow chart of treatment and imaging strategy. Pre-treatment and post-treatment imaging was performed by contrast-enhanced three-dimensional ultrasonography (CE 3D US) and CE 3D computed tomography (CT).
94
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
Fig. 2. A 70-year-old woman with hepatocellular carcinoma in the inferior posterior segment of the right liver lobe. (a–h) Contrast-enhanced three-dimensional ultrasonography images show that before radiofrequency (RF) ablation the HCC located adjacent to the right posterior portal vein was distinctly enhanced in the early phase, as shown on tomographic ultrasound images in plane A with a slice distance of 1.0 mm (a), plane B with a slice distance of 1.5 mm (b), and plane C with a slice distance of 1.0 mm (c) and on the sonographic angiogram (d) rendered by using maximum intensity mixed with surface mode. One day after treatment, adequate ablation in the absence of enhancement was detected as shown in the middle phase on the tomographic ultrasound images in plane A with a slice distance of 1.9 mm (e), plane B with a slice distance of 2.3 mm (f), and plane C with a slice distance of 2.4 mm (g), and on the sonographic angiogram (h) rendered by using average intensity mixed with surface mode. (i–p) Contrast-enhanced three-dimensional computed tomography images show that in the arterial phase before RF ablation the lesion was detected as an area of high attenuation on multiplanar images in the transverse plane (i), coronal plane (j) and sagittal plane (k), and on the CT angiogram (l) reconstructed by using the maximum-intensity-projection. One month after treatment adequate ablation was detected as an area of low attenuation in the arterial phase on multiplanar images in the transverse plane (m), coronal plane (n), and sagittal plane (o), and on the CT angiogram (p) reconstructed by using maximum-intensity-projection method. During a 20-month follow-up period, there was no local tumor progression detected on CE 3D CT. Arrowheads indicate the location of the original tumor before treatment or the ablated area after treatment.
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
95
Table 2 Evaluation by contrast-enhanced three-dimensional US and contrast-enhanced three-dimensional CT after radiofrequency ablation of HCCs. No. of lesions
One-day CE 3D US
One-month CE 3D CT
Follow-up CE 3D CT
n = 58 n=1 n=1 n=2 n=1
Adequate ablation Residual tumora Adequate ablation Residual tumora Residual tumora
Adequate ablation Adequate ablation Adequate ablation Residual tumorc Adequate ablation
No local tumor progression during follow-up period (range 11–21 months) Local tumor progression on 3-month CE 3D CTb Local tumor progression on 6-month CE 3D CTb No local tumor progression during 15-month follow-up
Note: “One-day CE 3D US” means contrast-enhanced three-dimensional ultrasonography performed 1 day after radiofrequency ablation (or the final ablation if multiple procedures were performed on the same lesion). “One-month CE 3D CT” and “follow-up CE 3D CT” mean contrast-enhanced three-dimensional computed tomography performed 1 month after radiofrequency ablation and that performed every 3 months after the RF ablation, respectively. a The lesion with residual tumors detected on 1-day CE 3D US, and the subject did not give their consent for the additional radiofrequency ablation. b The lesion with local tumor progression on the follow-up CE 3D CT was treated with the radiofrequency ablation or transarterial chemoembolization. c The lesion with residual tumors detected on 1-month CE 3D CT was then treated with the radiofrequency ablation.
early phase and middle phase, and was hypoechoic or isoechoic in the late phase of CE 3D US after ablation. When the lesion sites were evaluated on CE 3D CT after RF ablation, the images were evaluated as showing adequate ablation if after ablation a lowattenuation area was observed in the arterial phase, portal venous phase, and equilibrium phase covering the high-attenuation area seen in the arterial phase before ablation. Residual tumor on 1month CE 3D CT images was diagnosed if an area within the tumor showed high attenuation in the arterial phase and low-attenuation area in the portal venous phase, equilibrium phase, or both after ablation. Local tumor progression was evaluated if enhancement in the margin of the ablated areas in the arterial phase showed low attenuation in the portal venous phase, equilibrium phase, or both on the CE 3D CT images acquired every 3 months as follow-up examinations. 2.6. Statistical analysis Taken consideration of the findings on 1-month CE 3D CT images as a reference standard, the sensitivity, specificity, and accuracy of CE 3D US for detecting the adequate ablation was calculated. The concordance between the two modalities for detecting residual tumors was evaluated by the kappa test. Agreement was graded according to the kappa values in parentheses as poor (<0.20), moderate (0.20–0.40), fair (0.40–0.60), good (0.60–0.80), or excellent (0.80–1.0). The SPSS version 11.0 software package (SPSS, Tokyo, Japan) was used. 3. Results All 63 HCCs were treated by RF ablation and evaluated by using CE 3D US and CE 3D CT examinations before and after treatment. No subjects were excluded from the analysis during the follow-up period (range 3–21 months, mean, 15 months). Of the 63 HCCs, 54 were treated by a single RF ablation, and the other 9 HCCs were treated by 2 or 3 ablation procedures each to ensure adequate ablation. On the CE 3D US images before RF ablation all 63 HCCs appeared as hypervascular areas with diffused tumor enhancement in the early or middle phase (Fig. 2a–d) and as hypoechoic (n = 60) or isoechoic (n = 3) areas in the late phase, and on the CE 3D CT images they appeared as high-attenuation areas in the arterial phase (Fig. 2i–l) and as low-attenuation areas in the portal or equilibrium phase. Inter-reader variation for evaluating the effects of RF ablation was minimal. In 58 of the 63 lesions the two readers’ evaluation were the same. Different evaluations between two readers achieved in consensus in the other five lesions, including four evaluated as adequate ablation and one as residual tumors. The evaluation on CE 3D US and that on 1-month CE 3D CT achieved concordance in 61 lesions. As shown in Table 2, 59 of the 61 lesions showed no evidence of residual tumor in the ablated
areas on the CE 3D US images 1 day after the RF ablation or on the CE 3D CT images 1 month after RF ablation, and adequate ablation of 58 of the 59 lesions was confirmed by the subsequent follow-up CE 3D CT examinations. The absence of tumor vessels and tumor enhancement in the early phase, middle phase and late phase of the 1-day CE 3D US demonstrated the adequate ablation of HCCs (Fig. 2e–h), which was manifested as a low-attenuation area in the arterial phase, portal venous phase, and equilibrium phase of the CE 3D CT (Fig. 2m–p). In the remaining 2 of the 61 lesions, residual tumor due to inadequate ablation was detected on both 1-day CE 3D US and 1-month CE 3D CT. On 1-day CE 3D US residual tumor showed the hypervascular enhancement in the early phase and middle phase and hypoechoic in the late phase, however, additional RF ablation was not performed because the subjects did not give their consent. After 1 month, in both cases, residual tumor of high attenuation in the arterial phase and low attenuation in both the portal venous phase and equilibrium phases was detected in the peripheral ablated areas on the CE 3D CT. There was discordance between the evaluations by the two modalities in 2 of the 63 lesions. Residual tumor in both lesions was detected by CE 3D US 1 day after the RF ablation, and additional RF ablation was not performed because the subjects did not give their consent. The 1-month CE 3D CT showed no evidence of residual tumor in the ablated areas. Local tumor progression on one of these two lesions was identified during the follow-up period of CE 3D CT, but there was no evidence of tumor progression on the other lesion (Table 2). The kappa value for agreement between the two modalities was 0.65. Comparison of the results of 1-day CE 3D US with those of the 1-month CE 3D CT showed that the sensitivity of CE 3D US for detecting adequate ablation was 97% (59/61), specificity100% (2/2), and the accuracy 97% (61/63).
4. Discussion Advances in imaging techniques and contrast agents have improved the role of radiological imaging in evaluating the effect of RF ablation of HCCs [14–17]. On the dynamic radiological images, the necrotic areas after RF ablation of HCCs appeared as an absence of contrast agent perfusion, while residual portion of hypervascular tumors were usually enhanced in the vascular contrast phase and the contrast agent was washed out in the following parenchyma contrast phase [18–20]. CE CT is accepted as the gold standard for assessing whether HCC tumors are coagulated adequately, and it is performed routinely to evaluate the effect of ablation and for the long-term follow-up [21,22]. However, immediately after the procedure of RF ablation and during the early follow-up time, on CE CT images, the presence of hyperattenuation due to the dehydration produced by coagulative necrosis, and periablational enhancement due to reactive hyperemia, arteriovenous shunts, or fibrosis/giant
96
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97
cell reaction, hinder accurate use of this modality to evaluate RF ablation [23–25]. Compared to CE CT images, CE US images acquired in the early period after RF ablation are accompanied with few artifacts and have been proven to be useful for the early evaluation of RF ablation [26–28]. Wen et al. found that CE 2D US performed 5–7 days after RF ablation of 107 HCCs with the CHA mode and contrast agent Levovist had a sensitivity, specificity and accuracy of 95.3%, 100% and 98.1%, respectively, for detecting residual tumor [29]. In the present study, we used a newly developed imaging technique, CE 3D US, which has been reported to facilitate convenient and accurate observation of tortuous tumor vessels and tumor enhancement in a 3D view [9], to demonstrate residual tumor and thereby determine whether the tumor had been ablated completely or only partially. The reconstructed multiplanar TUI in the three orthogonal planes allowed the observers to detect different portions of tumors by synchronized presentation of multiple layers, while the sonographic angiograms spatially demonstrated the structures in the entire VOI with various view-angles. The 3D US images allowed the observers to compare the HCC tumors before treatment with the ablated areas after treatment in a 3D visualization. When the ablated areas and residual tumor were detected accurately in the three orthogonal planes of multiplanar images, sonographic angiograms could be used as a supplementary tool to view the lesion and surrounding anatomical structures in an angio-like perspective. CE 3D CT images were reported to be useful for preoperative visualization of the anatomic structures and for the intraoperative imaging guidance [30,31]. In our study CE 3D CT performed 1 month after RF ablation were regarded as the reference standard for the evaluation of RF ablation. Comparison with the results on 1-month CE 3D CT showed that CE 3D US had high sensitivity, specificity, and accuracy for detecting adequate ablation, thereby revealing a potential role for CE 3D US in assessing the effect of RF ablation. Unlike CT contrast media, the contrast agent Sonazoid used in our study does not disperse into the interstitial spaces. And with the assistance of CHA mode and high MI contrast conditions, which reduce microbubbles in microvessels but not in relatively large vessels, such as tumor vessels and portal veins [10], Sonazoid-enhanced 3D US facilitated elaborate observation of tumor vessels. Of the four lesions in which residual tumor was detected by 1-day CE 3D US, two were confirmed by 1-month CE 3D CT and then treated by additional RF ablation or transarterial chemoembolization, one was evaluated as adequate ablation by 1-month CE 3D CT and detected with local tumor progression in the marginal ablated areas by CE 3D CT after 3 months and then treated by additional RF ablation, and the remaining one was detected without local enhancement on 1month and the follow-up CE 3D CT. Our results to some extent imply CE 3D US is a sensitive tool for detecting residual tumor in the early time after RF ablation, although there may be some false-positive findings. Our study had some limitations. First, some lesions with residual tumor by 1-day CE 3D US after the initial treatment were subjected to the additional RF ablation to achieve adequate coagulation, and thus only four lesions with residual tumors after ablation were evaluated on CE 3D US, because these subjects did not consent to additional RF ablation. The small number of positive findings by CE 3D US and the relatively small study population may have affected the accuracy of assessment by CE 3D US. Second, one hypovascular HCC that was not enhanced on either imaging modality was excluded because it is difficult to evaluate the effect of RF ablation of hypovascular lesions on the basis of assessing tumor enhancement alone. Third, our initial study emphasized the local effect of RF ablation, and long-term follow-up to detect distant newly developed HCCs by CE 3D US needs further investigation.
In conclusion, because of its good concordance with the results of CE 3D CT, CE 3D US, which is convenient to perform and does not involve radiation exposure, can be used to evaluate the effects of RF ablation of HCCs with unique 3D view. References [1] Parkin DM, Bray F, Ferlay J, Pisani P. Estimating the world cancer burden: globocan 2000. Int J Cancer 2001;94:153–6. [2] Lau WY, Lai EC. Hepatocellular carcinoma: current management and recent advances. Hepatobiliary Pancreat Dis Int 2008;7:237–57. [3] El-Serag HB, Marrero JA, Rudolph L, Reddy KR. Diagnosis and treatment of hepatocellular carcinoma. Gastroenterology 2008;134:1752–63. [4] Veltri A, Moretto P, Doriguzzi A, Pagano E, Carrara G, Gandini G. Radiofrequency thermal ablation (RFA) after transarterial chemoembolization (TACE) as a combined therapy for unresectable non-early hepatocellular carcinoma (HCC). Eur Radiol 2006;16:661–9. [5] Howard JH, Tzeng CW, Smith JK, et al. Radiofrequency ablation for unresectable tumors of the liver. Am Surg 2008;74:594–600. [6] Crocetti L, Lencioni R. Thermal ablation of hepatocellular carcinoma. Cancer Imaging 2008;8:19–26. [7] Dietrich CF. 3D real time contrast enhanced ultrasonography—a new technique. Rofo 2002;174:160–3. [8] Ohto M, Kato H, Tsujii H, Maruyama H, Matsutani S, Yamagata H. Vascular flow patterns of hepatic tumors in contrast-enhanced 3-dimensional fusion ultrasonography using plane shift and opacity control modes. J Ultrasound Med 2005;24:49–57. [9] Luo W, Numata K, Morimoto M, et al. Clinical utility of contrast-enhanced three-dimensional ultrasound imaging with Sonazoid: findings on hepatocellular carcinoma lesions. Eur J Radiol. doi:10.1016/j.ejrad.2008.09.003 [accessed on October 17, 2008]. [10] Numata K, Morimoto M, Ogura T, et al. Ablation therapy guided by contrastenhanced sonography with Sonazoid for hepatocellular carcinoma lesions not detected by conventional sonography. J Ultrasound Med 2008;27:395–406. [11] Hatanaka K, Kudo M, Minami Y, et al. Differential diagnosis of hepatic tumors: value of contrast-enhanced harmonic sonography using the newly developed contrast agent, Sonazoid. Intervirology 2008;51 S1:61–9. [12] Nakano H, Ishida Y, Hatakeyama T, et al. Contrast-enhanced intraoperative ultrasonography equipped with late Kupffer-phase image obtained by sonazoid in patients with colorectal liver metastases. World J Gastroenterol 2008;14:3207–11. [13] Edey AJ, Ryan SM, Beese RC, Gordon P, Sidhu PS. Ultrasound imaging of liver metastases in the delayed parenchymal phase following administration of Sonazoid using a destructive mode technique (Agent Detection Imaging). Clin Radiol 2008;63:1112–20. [14] Buscarini E, Savoia A, Brambilla G, et al. Radiofrequency thermal ablation of liver tumors. Eur Radiol 2005;15:884–94. [15] Schima W, Ba-Ssalamah A, Kurtaran A, Schindl M, Gruenberger T. Posttreatment imaging of liver tumours. Cancer Imaging 2007;7:28–36. [16] 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. [17] McGahan JP, Khatri VP. Imaging findings after liver resection by using radiofrequency parenchymal coagulation devices: initial experiences. Radiology 2008;247:896–902. [18] Ding H, Kudo M, Onda H, et al. Evaluation of posttreatment response of hepatocellular carcinoma with contrast-enhanced coded phase-inversion harmonic US: comparison with dynamic CT. Radiology 2001;221:721–30. [19] Tsuda M, Majima K, Yamada T. Hepatocellular carcinoma after radiofrequency ablation therapy: dynamic CT evaluation of treatment. Clin Imaging 2001;25:409–15. [20] Fujioka C, Horiguchi J, Ishifuro M, et al. A feasibility study: evaluation of radiofrequency ablation therapy to hepatocellular carcinoma using image registration of preoperative and postoperative CT. Acad Radiol 2006;13:986–94. [21] Lim HS, Jeong YY, Kang HK, Kim JK, Park JG. Imaging features of hepatocellular carcinoma after transcatheter arterial chemoembolization and radiofrequency ablation. AJR Am J Roentgenol 2006;187:W341–9. [22] Geyik S, Akhan O, Abbaso˘glu O, et al. Radiofrequency ablation of unresectable hepatic tumors. Diagn Interv Radiol 2006;12:195–200. [23] Goldberg SN, Gazelle GS, Mueller PR. Thermal ablation therapy for focal malignancy: a unified approach to underlying principles, techniques, and diagnostic imaging guidance. AJR Am J Roentgenol 2000;174:323–31. [24] Dill-Macky MJ, Asch M, Burns P, Wilson S. Radiofrequency ablation of hepatocellular carcinoma: predicting success using contrast-enhanced sonography. AJR Am J Roentgenol 2006;186:S287–95. [25] Filippone A, Iezzi R, Di Fabio F, Cianci R, Grassedonio E, Storto ML. Multidetectorrow computed tomography of focal liver lesions treated by radiofrequency ablation: spectrum of findings at long-term follow-up. J Comput Assist Tomogr 2007;31:42–52. [26] Choi D, Lim HK, Lee WJ, et al. Radiofrequency ablation of liver cancer: early evaluation of therapeutic response with contrast-enhanced ultrasonography. Korean J Radiol 2004;5:185–98. [27] Kim CK, Choi D, Lim HK, et al. Therapeutic response assessment of percutaneous radiofrequency ablation for hepatocellular carcinoma: util-
W. Luo et al. / European Journal of Radiology 75 (2010) 91–97 ity of contrast-enhanced agent detection imaging. Eur J Radiol 2005;56: 66–73. [28] Vilana R, Bianchi L, Varela M, et al. Is microbubble-enhanced ultrasonography sufficient for assessment of response to percutaneous treatment in patients with early hepatocellular carcinoma? Eur Radiol 2006;16:2454–62. [29] Wen YL, Kudo M, Zheng RQ, et al. Radiofrequency ablation of hepatocellular carcinoma: therapeutic response using contrast-enhanced coded phase-inversion harmonic sonography. AJR Am J Roentgenol 2003;181:57–63.
97
[30] Endo I, Shimada H, Sugita M, et al. Role of three-dimensional imaging in operative planning for hilar cholangiocarcinoma. Surgery 2007;142: 666–75. [31] Dong Q, Xu W, Jiang B, et al. Clinical applications of computerized tomography 3-D reconstruction imaging for diagnosis and surgery in children with large liver tumors or tumors at the hepatic hilum. Pediatr Surg Int 2007;23: 1045–50.