Canadian Association of Radiologists Journal xx (2015) 1e8 www.carjonline.org
Vascular and Interventional Radiology
The Role of Contrast-Enhanced Ultrasound in Guiding Radiofrequency Ablation of Hepatocellular Carcinoma: A Retrospective Study Andy K. W. Chan, MD, FRCPCa,*, Chris Hegarty, MB, BCh, BAO, FRCPCa, Darren Klass, MD, PhD, FRCPCa, Eric Yoshida, MD, MHSc, FRCPCb, Stephen Chung, MD, PhD, FRCSCc, David M. Liu, MD, FRCPC, FSIRa, Stephen G. F. Ho, MD, FRCPCa, Alison C. Harris, MBChB, MRCP, FRCR, FRCPCa a
Department of Radiology, University of British Columbia, Faculty of Medicine, Vancouver General Hospital, Vancouver, British Columbia, Canada Department of Hepatology, University of British Columbia, Faculty of Medicine, Vancouver General Hospital, Vancouver, British Columbia, Canada c Department of Hepatobiliary Surgery, University of British Columbia, Faculty of Medicine, Vancouver General Hospital, Vancouver, British Columbia, Canada b
Abstract Purpose: The objective of the study was to determine the efficacy of contrast-enhanced ultrasound (CEUS) using ultrasound (US)-specific microbubbles in guiding radiofrequency ablation (RFA) of hepatocellular carcinoma (HCC). Methods: A retrospective analysis of 50 patients with HCC treated with CEUS guided RFA using perflutren at our institution was performed. CEUS images were first compared to B-mode US images performed at the same RFA session to determine the ability of CEUS to increase the conspicuity of lesions. A qualitative score (1 ¼ poor, 2 ¼ fair, 3 ¼ excellent) was used to grade the ability to visualize the lesions. The preprocedure CEUS images were then evaluated using the most recent prior contrast enhanced computed tomography (CT) or magnetic resonance imaging (MRI). The efficacy of the treatment was evaluated with short-term follow-up imaging (median 1 month) for presence of residual or recurrent disease. Results: CEUS allows at least fair visualization (score 2) in 78% (reader 1) and 80% (reader 2) of the lesions not visualized by B-mode US, and 50% (reader 1) and 42% (reader 2) of the lesions poorly visualized by B-mode US. Lesion appearances on CEUS are largely concordant with those on CT or MRI: 88% for reader 1, 96% for reader 2. With CEUS-guided RFA, complete response was achieved in the vast majority of the lesions at short-term follow-up: 82% for reader 1, 94% for reader 2. Conclusions: CEUS increases the conspicuity and provides better characterization of hypervascular HCC that are either not seen or poorly seen on B-mode US, and CEUS provides real-time guidance of RFA with good short-term treatment responses. Resume Objet : L’etude avait pour objectif de determiner si l’echographie avec injection d’un produit de contraste (compose de microbulles) permettait de guider efficacement l’ablation par radiofrequence d’un carcinome hepatocellulaire. Methodes : Notre analyse retrospective a porte sur 50 patients ayant subi une ablation de carcinome hepatocellulaire par radiofrequence sous echoguidage avec injection de perflutrene au sein de notre etablissement. Les images de l’echographie de contraste ont d’abord ete comparees aux images de l’echographie en mode B (echographie de brillance) realisee dans le cadre de la m^eme seance d’ablation par radiofrequence afin de determiner la capacite de l’echographie de contraste a accro^ıtre la perceptibilite des lesions. La perceptibilite des lesions a ete evaluee au moyen d’une note qualitative (1 ¼ faible, 2 ¼ passable, 3 ¼ excellente). Les images echographiques realisees avant la chirurgie ont ensuite ete analysees a l’aide d’images tirees du plus recent examen par tomodensitometrie (TD) ou d’imagerie par resonance magnetique (IRM) avec injection de produit de contraste. Enfin, l’efficacite du traitement a ete determinee a l’aide d’un examen de suivi a court terme (intervalle moyen d’un mois) visant a verifier la presence d’une affection residuelle ou recurrente. Resultats : L’echographie de contraste a permis de visualiser dans une mesure qualifiee d’au moins « passable » (note egale ou superieure a 2) 78 % (premiere lecture) et 80 % (deuxieme lecture) des lesions n’ayant pas ete discernees lors de l’echographie en mode B, et 50 % * Address for correspondence: Dr Andy K. W. Chan, MD, FRCPC, Department of Radiology, University of British Columbia, Vancouver General Hospital, Vancouver V5Z 1M9, Canada.
E-mail address: [email protected] (A. K. W. Chan).
0846-5371/$ - see front matter Ó 2015 Canadian Association of Radiologists. All rights reserved. http://dx.doi.org/10.1016/j.carj.2014.11.005
2
A. K. W. Chan et al. / Canadian Association of Radiologists Journal xx (2015) 1e8
(premiere lecture) et 42 % (deuxieme lecture) des lesions difficiles a discerner sur les images de l’echographie en mode B. L’aspect des lesions observees a l’echographie de contraste correspond la plupart du temps a l’aspect des lesions observees par TDM ou IRM (88 % a la premiere lecture, 96 % a la deuxieme). Dans la grande majorite des cas, l’ablation par radiofrequence guidee par echographie de contraste s’est traduite par une reponse therapeutique complete a l’examen de suivi a court terme (82 % a la premiere lecture et 94 % a la deuxieme). Conclusion : L’echographie de contraste accro^ıt la perceptibilite et renforce la caracterisation des carcinomes hepatocellulaires hypervascularises non visibles ou difficiles a discerner dans le cadre d’une echographie en mode B. Elle permet egalement le guidage en temps reel de l’ablation par radiofrequence, ce qui entra^ıne une bonne reponse therapeutique a court terme. Ó 2015 Canadian Association of Radiologists. All rights reserved. Key Words: Ultrasound; Contrast enhanced ultrasound; Radiofrequency ablation; Hepatocellular carcinoma; Gastrointestinal radiology; Interventional radiology
Hepatocellular carcinoma (HCC) is a common malignancy, with an increasing incidence due to hepatitis B and C viruses [1]. Chronic liver disease is the most important predisposing factor, accounting for 70%-90% of cases of HCC. Other common risk factors include alcohol, nonalcoholic fatty liver disease, tobacco, and diabetes [1]. Surgical resection or transplantation is the recommended treatment in patients with HCC and preserved liver functions. Unfortunately, less than 5% of cirrhotic patients with HCC fulfill these criteria [2]. Radiofrequency ablation (RFA) is a minimally invasive procedure that provides excellent local control of early stage HCC, allowing for lowered morbidity, thereby providing a treatment alternative for patients who are poor surgical candidates [3]. As a result, RFA has become an increasing important treatment modality for these patients. Image guidance plays a crucial role in RFA in the targeting of the tumour and assessing treatment response. Although contrast-enhanced computed tomography (CECT) provides excellent visualization, ultrasound (US) is often the modality of choice for guidance due to its realtime capabilities [4]. Conventional B-mode US, however, may not adequately visualize the tumour for treatment, and in the setting of previous ablation, identifying residual/ recurrent disease is often difficult [4e6]. With the introduction of second-generation microbubble US contrast agents, the sensitivity of contrast-enhanced US (CEUS) in detecting tumours is far superior to conventional B-mode US, and CEUS can be used to guide and monitor the RFA procedure to ensure successful treatment [6e19]. Finally, CEUS also has an important role in patients with renal impairment, where the uses of iodinated (CT) or gadolinium (magnetic resonance) contrast agents are relatively contraindicated. Multiple studies from Europe and Asia have demonstrated the ability of CEUS to characterize HCC lesions and guide RFA treatments [6e19]; however, our institution is the first in North America to routinely perform CEUS guided RFA clinically, due to our large population of patients with hepatitis B/C and cirrhosis. The purpose of this study is to determine the efficacy of CEUS in guiding RFA of HCC.
Methods Patient Selection and Study Design Institutional review board approval was from both the research ethics board of the university and health authority. This retrospective analysis was conducted between January 2011 and June 2012; patients treated with percutaneous RFA between Oct 2008 and May 2010 were included initially. The inclusion criteria used by our liver tumour group (consisting of radiologists, hepatobiliary surgeons, hepatologists, and oncologists) to select patients with HCC to be treated with CEUS guided RFA is based on the American Association for the Study of Liver Diseases Practice Guideline on the Management of Hepatocellular Carcinoma [20]: 1) A lesion is diagnosed radiologically as HCC if its size is greater than 1 cm, has the typical appearance of hypervascularity in the arterial phase and washout in portal venous or delay phases on contrast-enhanced CT or magnetic resonance imaging (MRI) examination, and is present in patients with underlying cirrhosis or hepatitis; 2) Patients with early stage disease, either solitary HCC or up to 3 nodules 3 cm in size, and not suitable for surgical resection or transplantation. Exclusion criteria includes: 1) vascular invasion; 2) subcapsular/subdiaphragmatic location, or adjacent to liver hilum, gallbladder, or gastrointestinal tract; 3) allergy to the CEUS contrast agent. Analyses of images from the selected patients were performed by 2 radiologists with up to 4 years’ experience in this field. Both readers were blinded to patient’s identifying information and any clinical history. The CEUS images were first compared to the conventional B-mode images performed on the same session to determine the ability of CEUS to increase the conspicuity of lesions for delivery for RFA, with a qualitative visualization score (1 ¼ poor, 2 ¼ fair, 3 ¼ excellent) used to grade the ability to visualize the lesion. The CEUS images were then evaluated against the most recent pretreatment cross-sectional imaging (contrastenhanced CT or MRI) as the gold standard, during which the enhancement pattern of the lesions on both modalities were compared for concordance. Finally, the primary outcome of the study is the efficacy of CEUS guided RFA treatment at
CEUS guided RFA treatment of HCC: a retrospective study / Canadian Association of Radiologists Journal xx (2015) 1e8
3
Figure 1. Utility of contrast-enhanced ultrasound (CEUS). Conventional B-mode US (left) demonstrates no definitive hepatic lesions in a background of heterogeneous hepatic parenchyma. CEUS (right) clearly demonstrates a hypervascular lesion during the arterial phase, consistent with HCC.
short-term (median 1 month) follow-up imaging (contrastenhanced CT or MRI), in which the presence or absence of disease is evaluated based on both the European Association for the Study of the Liver (EASL) and the modified Response Evaluation Criteria in Solid Tumors (modified RECIST) criteria [21,22]. Contrast Agent and CEUS Technique The second-generation microbubble US contrast agent used in our institution is perflutren (Definity, Lantheus Medical Imaging, North Billerica, MA). The agent is a phospholipid-coated microsphere with the size of a red blood cell and filled with perflurocarbon gas. Other secondgeneration agents used elsewhere in the world includes sulfur hexafluoride (SonoVue, Bracco, Milan, Italy) and perflubutane (Sonazoid, Daiichi Sankyo, Tokyo, Japan). All these agents persist in the bloodstream much longer than the first generation agents (Levovist; Schering AG, Berlin, Germany), allowing them to be used as a blood-pool agent in real-time scans (Figure 1). To optimize visualization of the tumour, patients are placed in either supine or left lateral decubitus positions. Low mechanical index (MI) is used to decrease power and preserve integrity of microbubbles. Furthermore, pulseinversion imaging allows suppression of echoes from tissue in favor of those from the bubbles. The vial is agitated prior to injection, which converts the clear contrast to the milky form. Subsequently, a bolus injection of 0.2-0.4 cc is
injected via an 18-gauge cannula at the antecubital vein, followed by 5-10 cc saline flush. The microbubbles are typically seen in 6-8 seconds following injection, and usually 2-3 injections are performed per patient, with a maximum of up to 1 vial. A CEUS liver study consists of 3 phases: 1) Arterial phase continuous cine loop images are obtained from 6-45 seconds after the injection of contrast; 2) Portal venous phase representative images are obtained from 45-90 seconds after the injection; 3) Delayed phase - representative images are obtained from 90 seconds up to 5 minutes after the injection. The typical appearance of classic HCC is similar to that of contrast-enhanced CT, with hypervascularity on arterial phase, and washout on portal venous and delayed phases (Figures 2-4). Iso/hypovascularity on arterial phase with slow or no washout on portal venous phase can be seen with variant HCC [23]. Recurrent or residual disease within previously ablated zone also demonstrates similar appearance (Figures 5 and 6). Percutaneous RFA Technique The electrode size is decided prior to the procedure based on the tumour size and the distance from the anticipated skin entry site to the tumour. Two RF ablation systems are utilized for treatment of liver tumours, LeVeen multitine electrode (Boston Scientific, Natick, MA) and the Cool-tip electrode (Covidien, Dublin, Ireland). The choice of electrode is based on operator preference and tumour location and is beyond the
Figure 2. Typical appearance of hepatocellular carcinoma on contrast-enhanced ultrasound. During arterial phase (left), the tumour demonstrates hypervascularity, with subsequent washout during portal venous and delayed phase imaging (right).
4
A. K. W. Chan et al. / Canadian Association of Radiologists Journal xx (2015) 1e8
Figure 3. Patient A: Triphasic liver computed tomography demonstrates a lesion in segment 4A of the liver, hypervascular on arterial phase (left image) with washout on both portal venous (middle image) and delayed phases (right image).
scope of this study. All ablations are performed by 1 of 3 interventional radiologists. Following adequate visualization with US, the skin entry site is cleaned and prepped and the site draped in a sterile fashion. Grounding pads provided by each of the vendors are placed on the upper thighs according to manufacturer recommendations. One percent Lidocaine is used to infiltrate the skin and under US guidance, the needle is advanced through the subcutaneous tissues to the liver capsule and further local anesthetic administered. Sedo-analgesia is then administered by a trained nurse intravenously and titrated throughout the procedure. A combination of Midazolam and Fentanyl are administered intravenously and continuous monitoring of blood pressure, pulse, oxygen saturation and respiratory rate performed by a trained nurse. A small incision is made using a no. 11 scalpel blade and under US guidance, the probe is inserted into the lesion and in the case of the LeVeen multitine needle, the tines deployed. The ablations are performed according to the manufacturer recommendations, in the case of the LeVeen multitine electrode, a specific algorithm is followed according to the size of the electrode chosen for the procedure and continued until satisfactory roll-off is achieved. In the case of the Cool-tip electrode, energy is delivered into the tumour for a total of 12 minutes and the procedure stopped regardless of whether roll-off was achieved.
If deemed necessary by the operator, further ablations are performed in order to achieve a minimum of 5 mm of healthy ablated tissue around the tumour. It is difficult to accurately assess the exact extent of the ablation zone and this is estimated based on the size of the tumour and the needle used for ablation. Following the ablation, the needle tract is ablated and the electrode removed. Results Pretreatment Imaging Prior to receiving their CEUS-guided RFA treatments, the patient’s pretreatment imaging (CT or MRI) was first reviewed by the 2 radiologists. Reader 1 reviewed 50 lesions, while reader 2 reviewed 47 lesions. A small number of patients (2 patients) had decreased renal function and therefore no iodinated or gadolinium based contrast agent was administered. Among the patients that received contrast, almost all lesions demonstrate the typical characteristics of hepatocellular carcinoma: hypervascular on arterial phase with subsequent washout on portal venous/delayed phase imaging. The characteristics of these lesions are summarized in Table 1. CEUS during Intervention
Figure 4. Patient A: Contrast-enhanced ultrasound (CEUS) performed 1 month later confirms a hypervascular lesion in segment 4A of the liver (Figure 3). Radiofrequency ablation was then performed under CEUS guidance.
Table 2 demonstrates the analysis of the lesions from patients who underwent CEUS-guided RFA. Reader 1 analyzed 50 lesions, 41 of these lesions were visualized with B-mode US and 37 lesions visualized with CEUS. Reader 2 analyzed 32 lesions, 27 of these lesions were visualized with B-mode US and 31 lesions visualized with CEUS. There is moderate interobserver agreement for both the B-mode and CEUS scoring, with kappa values of 0.455 and 0.543, respectively. Of the lesions visualized, the visualization score of lesions with CEUS is 2.3 0.7 for reader 1 and 2.4 0.8 for reader 2, while the visualization score of lesions with B-mode US is 1.7 0/7 for reader 1 and 1.8 0.8 for reader 2. The differences between the CEUS and B-mode are not statistically significant for either reader (P value of 0.734 for reader 1 and 0.950 for reader 2). Subjectively, however, reader 1 found CEUS was helpful compared to B-mode US alone in 70.3% of lesions, whereas reader 2 found CEUS helpful in 71.0% of lesions.
CEUS guided RFA treatment of HCC: a retrospective study / Canadian Association of Radiologists Journal xx (2015) 1e8
5
Figure 5. Patient A: Triphasic liver computed tomography at 6-month follow up demonstrates a hypervascular nodular focus within the segment 4A ablation zone, in keeping with recurrent disease (left image ¼ arterial phase; middle image ¼ portal venous phase; right image ¼ delayed phase).
Table 3 demonstrates the distribution of B-mode and CEUS visualization scores, while Table 4 compares the CEUS visualization scores to those of B-mode scores. Of the lesions not visualized by B-mode US, CEUS allows at least fair visualization (i.e. score 2) for 78% (7 of 9) and 80% (4 of 5) of the lesions by reader 1 and reader 2, respectively. Similarly, of the lesions poorly visualized by B-mode US (ie, score of 1), CEUS allows at least fair visualization (ie, score 2) for 50% (9 of 18) and 42% (5 of 12) of the lesions by reader 1 and reader 2, respectively. These data suggests that CEUS is more valuable for evaluation of lesions that are either not visualized or poorly visualized by B-mode US. Follow-Up Imaging Table 5 demonstrates the outcome of CEUS-guided RFA treatment on follow-up imaging. The vast majority of the follow-up was performed 1 month after the treatment, with a total number of 7 cases performed from 2-7 months after and 1 case performed 13 months after the treatment. In terms of imaging modality, almost all cases were performed with CT, except 1 case with MRI and 1 case with CEUS. Complete response (CR) was achieved in the vast majority of the lesions: 37 of 45 lesions (82.2% with 95% confidence interval of 11.2%) for reader 1, and 30 of 32 lesions (93.8% with 95% confidence interval of 8.4%) for reader 2, suggesting
CEUS-guided RFA is an efficacious treatment modality for patient with HCC at least in the short-term setting. Discussion RFA has become increasingly utilized in the treatment of HCC, due to it being minimally invasive, with lower morbidity rates compared to surgery [3]. Conventional B-mode US has been the standard modality used for guidance of probe placement. Unfortunately, visualization of the target tumour may not be adequate, and the identification of residual/recurrent disease is often difficult in the setting of previous ablation [4e6]. Since the introduction of secondgeneration microbubble US contrast agents, multiple studies have investigated the various roles of CEUS in management of patients with HCC. Several studies have confirmed the ability of CEUS to detect and characterize HCC [5e7,9e12,14e19,24e26]. Giorgio et al. [13] reported that the appearance of the lesions was concordant between CEUS and CT in 82.4% (61 of 74) of the cases, and there were no statistically significant differences in the visualization of the hypervascularity patterns of HCC between the 2 modalities. For detection of local recurrent disease, the morphologic patterns on CEUS are concordant with those on CT in all 17 cases of the study by Miyamoto et al. [5], and subsequently they successfully performed RFA in all of these recurrent cases. Table 1 Lesions characteristics on pretreatment imaging
Figure 6. Patient A: Contrast-enhanced ultrasound performed 1 week after the 6-month follow-up computed tomography (CT) (Figure 5) demonstrates the segment 4A recurrent disease, which closely resembles the appearance on CT (Figure 5).
Number of lesions Imaging modality CT MR Enhancement pattern No contrast administered 1 ¼ Hypervascular 2 ¼ Isovascular 3 ¼ Hypovascular Lesion type New lesion Residual disease (and pattern) 1 ¼ Peripheral 2 ¼ Central 3 ¼ Circumferential 4 ¼ Satellite
Reader 1
Reader 2
50
47
45 5
42 5
3 46 0 1
3 43 1 0
36 11 9 2 0 0
29 15 8 7 0 0
CT ¼ computed tomography; MR ¼ magnetic resonance.
6
A. K. W. Chan et al. / Canadian Association of Radiologists Journal xx (2015) 1e8
Table 2 Analysis of lesions from patients underwent CEUS-guided RFA
Number of lesions analyzed Number of lesions visualized with B-mode US B-mode score (mean SD) Number of lesions visualized with CEUS CEUS score (mean SD) Number of lesions where CEUS is helpful compared to B-mode alone Number of lesions where CEUS and pretreatment contrast-enhanced CT/MR are both available Number of lesions where CEUS is concordant with pretreatment CT/MR
Table 4 Comparison of B-mode and CEUS visualization scores
Reader 1
Reader 2
50 41 1.7 0.7 37 2.3 0.7 26 (70.3%)
32 27 1.8 0.8 31 2.4 0.8 22 (71.0%)
34
24
CEUS 30 (88.2%)
23 (95.9%)
Worse Equal Better Total
Not visualized
1
2
3
Total
N/A 2 7 9
7 2 9 18
5 7 5 17
3 3 N/A 6
15 14 21
Worse Equal Better Total
Not visualized
1
2
3
Total
N/A 1 4 5
6 1 5 12
3 2 4 9
2 4 N/A 6
11 8 13
CEUS ¼ contrast-enhanced ultrasound.
Not visualized
1
2
3
Total
superior to CEUS either, with Vilana et al. [25] reporting a sensitivity of only 20% (2 of 10 residual nodules detected). CEUS has proven to be much more accurate at evaluating treatment response at 1-month follow-up. Using contrastenhanced CT as the gold standard, the sensitivity/specificity of CEUS reported are: 91%/97% specificity (41 patients) by Vilana et al. [25], 92.3%/100% (100 patients) by Ricci et al [18], 83.3%/100% (110 nodules) by Salvaggio et al. [16], and 92.9%/98.8% by Bartolotta et al. [7]. Another 1-month follow-up study performed by Lu et al. [10] compared the diagnostic performance of both conventional B-mode US and CEUS to that of contrast-enhanced CT/MRI in 139 patients, in which the sensitivity/accuracy for B-mode US and CEUS are 78.2%/76.3% versus 97.0%/94.2%, respectively. With multiple studies demonstrating the ability of CEUS to detect and characterize HCC, CEUS can be used to guide RFA treatment and ultimately improve patient outcome. Minami et al. [6] retrospectively analyzed 66 patients with 108 hepatic malignancies (HCC and liver metastasis), in which technically successful ablation was achieved in 94% of patients in the first session. The unsuccessful cases were attributed to either deep tumour location, location behind costal bone, or insufficient enhancement. None of these patients demonstrated local tumour progression during follow-up imaging ranging from 1-12 months, although 5 patients from each of the HCC and metastasis groups demonstrated distant metastases in the liver. Maruyama et al. [4] conducted a prospective study consists of 55
2 0 4 3 9
7 2 3 6 18
2 3 7 5 17
1 1 1 3 6
12 6 15 17
Table 5 Outcome on short-term follow-up imaging
Not visualized
1
2
3
Total
1 0 1 3 5
6 1 2 3 12
1 2 2 4 9
0 1 1 4 6
8 4 6 14
Table 3 Distribution of B-mode and CEUS visualization scores B-mode
Not visualized 1 2 3 Total
CEUS
Reader 2
The use of CEUS immediately posttreatment is potentially of great benefit to the patients, since the detection of incomplete response would enable the delivery of additional treatment in the same session [7,26,27]. Dill-Macky et al. [26] reported a sensitivity of 88% and specificity of 40% in the ability of postprocedural CEUS to predict 2-week follow-up CT/MRI results in 21 lesions. The low specificity is likely related to the fact that it may be difficult to differentiate reactive hyperemia (either from tissue response or arteriovenous shunting) from residual tumour. It has been suggested that reactive hyperemia is usually uniform in thickness and surrounds the ablation zone in a rind-like fashion [27], whereas residual disease is often nodular or crescentic in shape. Evaluating treatment response at 24-hour posttreatment is also suboptimal, with Vilana et al. [25] reporting a sensitivity of 27.3% (3 of 11 residual nodules detected) when compared to contrast-enhanced CT at 1-month follow-up. This is due to obscuration of nodules from the presence of gas with strong posterior echoes. The use of contrast-enhanced CT at 24-hour posttreatment is not
CEUS
Reader 1
B-mode
CEUS ¼ contrast-enhanced ultrasound; CT ¼ computed tomography; MR ¼ magnetic resonance; RFA ¼ radiofrequency ablation; US ¼ ultrasound.
Reader 1
B-mode
Reader 1
B-mode Reader 2 CEUS
Not visualized 1 2 3 Total
CEUS ¼ contrast-enhanced ultrasound.
Complete response (CR) Partial response (PR) Stable disease (SD) Progressive disease (PD) Total
Reader 2
EASL
RECIST
EASL
RECIST
37 5 1 2 45
37 5 1 2 45
30 2 0 0 32
30 2 0 0 32
EASL ¼ European Association for the Study of RECIST ¼ Response Evaluation Criteria in Solid Tumors.
the
Liver;
CEUS guided RFA treatment of HCC: a retrospective study / Canadian Association of Radiologists Journal xx (2015) 1e8
hypervascular lesions (from 44 cirrhotic patients) diagnosed on contrast-enhanced CT but not visualized by conventional B-mode US. Among these 55 lesions, 53 were identified by CEUS, and 42 lesions were treated percutaneously with either RFA or ethanol ablation. All of these treated hypervascular lesions become hypovascular on posttreatment CT. As a result, they concluded that CEUS improves the detectability of lesions that are occult on conventional US and allows wider application of percutaneous treatment in cirrhotic patients. To our knowledge, this is the first study in North America to investigate the role of CEUS for guiding RFA treatment. The results from our study are largely concordant with those from the studies described above. First, we confirmed the ability of CEUS to increase conspicuity of lesions compared to B-mode US in general (mean visualization score of 2.3 0.7 vs 1.7 0.7 for reader 1 and 2.4 0.8 vs 1.8 0.8 for reader 2). However, the difference in visualization is more significant for lesions that are either not visualized or poorly visualized with B-mode US. For lesions not visualized by B-mode US, CEUS allows at least fair visualization (ie, score 2) for 78% (7 of 9) and 80% (4 of 5) of the lesions by reader 1 and reader 2, respectively. For lesions poorly visualized by B-mode US (ie, score of 1), CEUS allows at least fair visualization (ie, score 2) for 50% (9 of 18) and 42% (5 of 12) of the lesions by reader 1 and reader 2, respectively. The appearances on CEUS is largely concordant with those on contrast-enhanced CT or MRI (30 of 34 lesions ¼ 88% for reader 1; 23 of 24 lesions ¼ 96% for reader 2). The authors acknowledge limitations of this study: 1. This is small cohort of patients, and large studies would provide more robust data. Our data, however, support the short-term efficacy of CEUS as an adjunct to conventional B-mode US for guiding RFA treatments based on the EASL and RECIST criteria. 2. It is a retrospective study and a control group consisting of patients treated under the guidance of a modality that is currently used in most clinical settings (conventional B-mode US, contrast-enhanced CT), was not possible due to clinical constraints. As a result, although we were able to document the efficacy of CEUS guided RFA at 1-month follow-up in terms of the rate of residual or recurrent disease, superiority or equivalence compared to these modalities cannot be evaluated. To address this issue, we aim to perform a retrospective study of patients treated with RFA before CEUS has been implemented in our institution, thus allowing the analysis of outcomes in patients treated under the guidance of conventional B-mode US. One month is a short follow-up interval, and collection of longer-term outcome data (eg, 5 years) by evaluating subsequently follow-up imaging will be the next stage of the study. Other outcome variables such as mortality or disease-free survival could also be analyzed to further strengthen the study.
7
Conclusion CEUS increases tumour conspicuity and provides better characterization of hypervascular lesions that are either not seen or poorly seen on conventional B-mode US and provides real-time guidance of RFA with good short-term treatment responses. References [1] Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet 2012; 379:1245e55. [2] Bruix J, Llovet JM. Prognostic prediction and treatment strategy in hepatocellular carcinoma. Hepatology 2002;35:519e24. [3] Lencioni R, Cioni D, Crocetti L, et al. Early-Stage hepatocellular carcinoma in patients with cirrhosis: long-term results of percutaneous image-guided radiofrequency ablation. Radiology 2005;234:961e7. [4] Maruyama H, Takahashi M, Ishibashi H, et al. Ultrasound-guided treatments under low acoustic power contrast harmonic imaging for hepatocellular carcinomas undetected by B-mode ultrasonography. Liver Int 2009;29:708e14. [5] Miyamoto N, Hiramatsu K, Tsuchiya K, et al. Contrast-enhanced sonography-guided radiofrequency ablation for the local recurrence of previously treated hepatocellular carcinoma undetected by B-mode sonography. J Clin Ultrasound 2010;38:339e45. [6] Minami Y, Kudo M, Hatanaka K, et al. Radiofrequency ablation guided by contrast harmonic sonography using perfluorocarbon microbubbles (Sonazoid) for hepatic malignancies: an initial experience. Liver Int 2010;30:759e64. [7] Bartolotta TV, Taibbi A, Midiri M, et al. Hepatocellular cancer response to radiofrequency tumor ablation: contrast-enhanced ultrasound. Abdom Imaging 2007;33:501e11. [8] Shiozawa K, Watanabe M, Takayama R, et al. Evaluation of local recurrence after treatment for hepatocellular carcinoma by contrastenhanced ultrasonography using Sonazoid: comparison with dynamic computed tomography. J Clin Ultrasound 2010;38:182e9. [9] Koda M, Mandai M, Matono T, et al. Assessment of the ablated area after radiofrequency ablation by contrast-enhanced sonography; comparison with virtual sonography with magnetic navigation. Clin Imaging 2010;34:60e4. [10] Lu M-D, Yu X-L, Li A-H, et al. Comparison of contrast enhanced ultrasound and contrast enhanced CT or MRI in monitoring percutaneous thermal ablation procedure in patients with hepatocellular carcinoma: a multi-center study in China. Ultrasound Med Biol 2007;33:1736e49. [11] Sparchez Z, Radu P, Anton O, et al. Contrast enhanced ultrasound in assessing therapeutic response in ablative treatments of hepatocellular carcinoma. J Gastrointestin Liver Dis 2009;18:243e8. [12] Leen E, Kumar S, Khan SA, et al. Contrast-enhanced 3D ultrasound in the radiofrequency ablation of liver tumors. World J Gastroenterol 2009;15:289e99. [13] Giorgio A, Ferraioli G, Tarantino L, et al. Contrast-enhanced sonographic appearance of hepatocellular carcinoma in patients with cirrhosis: comparison with contrast-enhanced helical CT appearance. AJR Am J Roentgenol 2004;183:1319e26. [14] Kisaka Y, Hirooka M, Koizumi Y, et al. Contrast-enhanced sonography with abdominal virtual sonography in monitoring radiofrequency ablation of hepatocellular carcinoma. J Clin Ultrasound 2009;38:138e44. [15] Gallotti A, D’Onofrio M, Ruzzenente A, et al. Contrast-enhanced ultrasonography (CEUS) immediately after percutaneous ablation of hepatocellular carcinoma. Radiol Med 2009;114:1094e105. [16] Salvaggio G, Campisi A, Greco V, et al. Evaluation of posttreatment response of hepatocellular carcinoma: comparison of ultrasonography with second-generation ultrasound contrast agent and multidetector CT. Abdom Imaging 2009;35:447e53. [17] Numata K, Fukuda H, Ohto M, et al. Evaluation of the therapeutic efficacy of high-intensity focused ultrasound ablation of hepatocellular
8
[18]
[19]
[20]
[21]
[22]
A. K. W. Chan et al. / Canadian Association of Radiologists Journal xx (2015) 1e8 carcinoma by three-dimensional sonography with a perflubutane-based contrast agent. Eur J Radiol 2010;75:e67e75. Ricci P, Cantisani V, Drudi F, et al. Is contrast-enhanced US alternative to spiral CT in the assessment of treatment outcome of radiofrequency ablation in hepatocellular carcinoma? Ultraschall Med 2009;30:252e8. Luo W, Numata K, Morimoto M, et al. Role of Sonazoid-enhanced threedimensional ultrasonography in the evaluation of percutaneous radiofrequency ablation of hepatocellular carcinoma. Eur J Radiol 2010;75:91e7. Bruix J, Sherman M. American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma: an update. Hepatology 2011;53:1020e2. Riaz A, Memon K, Miller FH, et al. Role of the EASL, RECIST, and WHO response guidelines alone or in combination for hepatocellular carcinoma: radiologic-pathologic correlation. J Hepatol 2011;54:695e704. Forner A, Ayuso C, Varela M, et al. Evaluation of tumor response after locoregional therapies in hepatocellular carcinoma. Cancer 2009;115: 616e23.
[23] Jang H-J, Kim TK, Burns PN, et al. Enhancement patterns of hepatocellular carcinoma at contrast-enhanced US: comparison with histologic differentiation. Radiology 2007;244:898e906. [24] Luo W, Numata K, Morimoto M, et al. Clinical utility of contrastenhanced three-dimensional ultrasound imaging with Sonazoid: Findings on hepatocellular carcinoma lesions. Eur J Radiol 2009;72: 425e31. [25] 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;16:2454e62. [26] Dill-Macky MJ, Asch M, Burns P, et al. Radiofrequency ablation of hepatocellular carcinoma: predicting success using contrast-enhanced sonography. AJR Am J Roentgenol 2006;186:S287e95. [27] Solbiati L, Ierace T, Tonolini M, et al. Guidance and monitoring of radiofrequency liver tumor ablation with contrast-enhanced ultrasound. Eur J Radiol 2004;51(Suppl):S19e23.
Vascular and Interventional Radiology
The Role of Contrast-Enhanced Ultrasound in Guiding Radiofrequency Ablation of Hepatocellular Carcinoma: A Retrospective Study Andy K. W. Chan, MD, FRCPCa,*, Chris Hegarty, MB, BCh, BAO, FRCPCa, Darren Klass, MD, PhD, FRCPCa, Eric Yoshida, MD, MHSc, FRCPCb, Stephen Chung, MD, PhD, FRCSCc, David M. Liu, MD, FRCPC, FSIRa, Stephen G. F. Ho, MD, FRCPCa, Alison C. Harris, MBChB, MRCP, FRCR, FRCPCa a
Department of Radiology, University of British Columbia, Faculty of Medicine, Vancouver General Hospital, Vancouver, British Columbia, Canada Department of Hepatology, University of British Columbia, Faculty of Medicine, Vancouver General Hospital, Vancouver, British Columbia, Canada c Department of Hepatobiliary Surgery, University of British Columbia, Faculty of Medicine, Vancouver General Hospital, Vancouver, British Columbia, Canada b
Abstract Purpose: The objective of the study was to determine the efficacy of contrast-enhanced ultrasound (CEUS) using ultrasound (US)-specific microbubbles in guiding radiofrequency ablation (RFA) of hepatocellular carcinoma (HCC). Methods: A retrospective analysis of 50 patients with HCC treated with CEUS guided RFA using perflutren at our institution was performed. CEUS images were first compared to B-mode US images performed at the same RFA session to determine the ability of CEUS to increase the conspicuity of lesions. A qualitative score (1 ¼ poor, 2 ¼ fair, 3 ¼ excellent) was used to grade the ability to visualize the lesions. The preprocedure CEUS images were then evaluated using the most recent prior contrast enhanced computed tomography (CT) or magnetic resonance imaging (MRI). The efficacy of the treatment was evaluated with short-term follow-up imaging (median 1 month) for presence of residual or recurrent disease. Results: CEUS allows at least fair visualization (score 2) in 78% (reader 1) and 80% (reader 2) of the lesions not visualized by B-mode US, and 50% (reader 1) and 42% (reader 2) of the lesions poorly visualized by B-mode US. Lesion appearances on CEUS are largely concordant with those on CT or MRI: 88% for reader 1, 96% for reader 2. With CEUS-guided RFA, complete response was achieved in the vast majority of the lesions at short-term follow-up: 82% for reader 1, 94% for reader 2. Conclusions: CEUS increases the conspicuity and provides better characterization of hypervascular HCC that are either not seen or poorly seen on B-mode US, and CEUS provides real-time guidance of RFA with good short-term treatment responses. Resume Objet : L’etude avait pour objectif de determiner si l’echographie avec injection d’un produit de contraste (compose de microbulles) permettait de guider efficacement l’ablation par radiofrequence d’un carcinome hepatocellulaire. Methodes : Notre analyse retrospective a porte sur 50 patients ayant subi une ablation de carcinome hepatocellulaire par radiofrequence sous echoguidage avec injection de perflutrene au sein de notre etablissement. Les images de l’echographie de contraste ont d’abord ete comparees aux images de l’echographie en mode B (echographie de brillance) realisee dans le cadre de la m^eme seance d’ablation par radiofrequence afin de determiner la capacite de l’echographie de contraste a accro^ıtre la perceptibilite des lesions. La perceptibilite des lesions a ete evaluee au moyen d’une note qualitative (1 ¼ faible, 2 ¼ passable, 3 ¼ excellente). Les images echographiques realisees avant la chirurgie ont ensuite ete analysees a l’aide d’images tirees du plus recent examen par tomodensitometrie (TD) ou d’imagerie par resonance magnetique (IRM) avec injection de produit de contraste. Enfin, l’efficacite du traitement a ete determinee a l’aide d’un examen de suivi a court terme (intervalle moyen d’un mois) visant a verifier la presence d’une affection residuelle ou recurrente. Resultats : L’echographie de contraste a permis de visualiser dans une mesure qualifiee d’au moins « passable » (note egale ou superieure a 2) 78 % (premiere lecture) et 80 % (deuxieme lecture) des lesions n’ayant pas ete discernees lors de l’echographie en mode B, et 50 % * Address for correspondence: Dr Andy K. W. Chan, MD, FRCPC, Department of Radiology, University of British Columbia, Vancouver General Hospital, Vancouver V5Z 1M9, Canada.
E-mail address: [email protected] (A. K. W. Chan).
0846-5371/$ - see front matter Ó 2015 Canadian Association of Radiologists. All rights reserved. http://dx.doi.org/10.1016/j.carj.2014.11.005
2
A. K. W. Chan et al. / Canadian Association of Radiologists Journal xx (2015) 1e8
(premiere lecture) et 42 % (deuxieme lecture) des lesions difficiles a discerner sur les images de l’echographie en mode B. L’aspect des lesions observees a l’echographie de contraste correspond la plupart du temps a l’aspect des lesions observees par TDM ou IRM (88 % a la premiere lecture, 96 % a la deuxieme). Dans la grande majorite des cas, l’ablation par radiofrequence guidee par echographie de contraste s’est traduite par une reponse therapeutique complete a l’examen de suivi a court terme (82 % a la premiere lecture et 94 % a la deuxieme). Conclusion : L’echographie de contraste accro^ıt la perceptibilite et renforce la caracterisation des carcinomes hepatocellulaires hypervascularises non visibles ou difficiles a discerner dans le cadre d’une echographie en mode B. Elle permet egalement le guidage en temps reel de l’ablation par radiofrequence, ce qui entra^ıne une bonne reponse therapeutique a court terme. Ó 2015 Canadian Association of Radiologists. All rights reserved. Key Words: Ultrasound; Contrast enhanced ultrasound; Radiofrequency ablation; Hepatocellular carcinoma; Gastrointestinal radiology; Interventional radiology
Hepatocellular carcinoma (HCC) is a common malignancy, with an increasing incidence due to hepatitis B and C viruses [1]. Chronic liver disease is the most important predisposing factor, accounting for 70%-90% of cases of HCC. Other common risk factors include alcohol, nonalcoholic fatty liver disease, tobacco, and diabetes [1]. Surgical resection or transplantation is the recommended treatment in patients with HCC and preserved liver functions. Unfortunately, less than 5% of cirrhotic patients with HCC fulfill these criteria [2]. Radiofrequency ablation (RFA) is a minimally invasive procedure that provides excellent local control of early stage HCC, allowing for lowered morbidity, thereby providing a treatment alternative for patients who are poor surgical candidates [3]. As a result, RFA has become an increasing important treatment modality for these patients. Image guidance plays a crucial role in RFA in the targeting of the tumour and assessing treatment response. Although contrast-enhanced computed tomography (CECT) provides excellent visualization, ultrasound (US) is often the modality of choice for guidance due to its realtime capabilities [4]. Conventional B-mode US, however, may not adequately visualize the tumour for treatment, and in the setting of previous ablation, identifying residual/ recurrent disease is often difficult [4e6]. With the introduction of second-generation microbubble US contrast agents, the sensitivity of contrast-enhanced US (CEUS) in detecting tumours is far superior to conventional B-mode US, and CEUS can be used to guide and monitor the RFA procedure to ensure successful treatment [6e19]. Finally, CEUS also has an important role in patients with renal impairment, where the uses of iodinated (CT) or gadolinium (magnetic resonance) contrast agents are relatively contraindicated. Multiple studies from Europe and Asia have demonstrated the ability of CEUS to characterize HCC lesions and guide RFA treatments [6e19]; however, our institution is the first in North America to routinely perform CEUS guided RFA clinically, due to our large population of patients with hepatitis B/C and cirrhosis. The purpose of this study is to determine the efficacy of CEUS in guiding RFA of HCC.
Methods Patient Selection and Study Design Institutional review board approval was from both the research ethics board of the university and health authority. This retrospective analysis was conducted between January 2011 and June 2012; patients treated with percutaneous RFA between Oct 2008 and May 2010 were included initially. The inclusion criteria used by our liver tumour group (consisting of radiologists, hepatobiliary surgeons, hepatologists, and oncologists) to select patients with HCC to be treated with CEUS guided RFA is based on the American Association for the Study of Liver Diseases Practice Guideline on the Management of Hepatocellular Carcinoma [20]: 1) A lesion is diagnosed radiologically as HCC if its size is greater than 1 cm, has the typical appearance of hypervascularity in the arterial phase and washout in portal venous or delay phases on contrast-enhanced CT or magnetic resonance imaging (MRI) examination, and is present in patients with underlying cirrhosis or hepatitis; 2) Patients with early stage disease, either solitary HCC or up to 3 nodules 3 cm in size, and not suitable for surgical resection or transplantation. Exclusion criteria includes: 1) vascular invasion; 2) subcapsular/subdiaphragmatic location, or adjacent to liver hilum, gallbladder, or gastrointestinal tract; 3) allergy to the CEUS contrast agent. Analyses of images from the selected patients were performed by 2 radiologists with up to 4 years’ experience in this field. Both readers were blinded to patient’s identifying information and any clinical history. The CEUS images were first compared to the conventional B-mode images performed on the same session to determine the ability of CEUS to increase the conspicuity of lesions for delivery for RFA, with a qualitative visualization score (1 ¼ poor, 2 ¼ fair, 3 ¼ excellent) used to grade the ability to visualize the lesion. The CEUS images were then evaluated against the most recent pretreatment cross-sectional imaging (contrastenhanced CT or MRI) as the gold standard, during which the enhancement pattern of the lesions on both modalities were compared for concordance. Finally, the primary outcome of the study is the efficacy of CEUS guided RFA treatment at
CEUS guided RFA treatment of HCC: a retrospective study / Canadian Association of Radiologists Journal xx (2015) 1e8
3
Figure 1. Utility of contrast-enhanced ultrasound (CEUS). Conventional B-mode US (left) demonstrates no definitive hepatic lesions in a background of heterogeneous hepatic parenchyma. CEUS (right) clearly demonstrates a hypervascular lesion during the arterial phase, consistent with HCC.
short-term (median 1 month) follow-up imaging (contrastenhanced CT or MRI), in which the presence or absence of disease is evaluated based on both the European Association for the Study of the Liver (EASL) and the modified Response Evaluation Criteria in Solid Tumors (modified RECIST) criteria [21,22]. Contrast Agent and CEUS Technique The second-generation microbubble US contrast agent used in our institution is perflutren (Definity, Lantheus Medical Imaging, North Billerica, MA). The agent is a phospholipid-coated microsphere with the size of a red blood cell and filled with perflurocarbon gas. Other secondgeneration agents used elsewhere in the world includes sulfur hexafluoride (SonoVue, Bracco, Milan, Italy) and perflubutane (Sonazoid, Daiichi Sankyo, Tokyo, Japan). All these agents persist in the bloodstream much longer than the first generation agents (Levovist; Schering AG, Berlin, Germany), allowing them to be used as a blood-pool agent in real-time scans (Figure 1). To optimize visualization of the tumour, patients are placed in either supine or left lateral decubitus positions. Low mechanical index (MI) is used to decrease power and preserve integrity of microbubbles. Furthermore, pulseinversion imaging allows suppression of echoes from tissue in favor of those from the bubbles. The vial is agitated prior to injection, which converts the clear contrast to the milky form. Subsequently, a bolus injection of 0.2-0.4 cc is
injected via an 18-gauge cannula at the antecubital vein, followed by 5-10 cc saline flush. The microbubbles are typically seen in 6-8 seconds following injection, and usually 2-3 injections are performed per patient, with a maximum of up to 1 vial. A CEUS liver study consists of 3 phases: 1) Arterial phase continuous cine loop images are obtained from 6-45 seconds after the injection of contrast; 2) Portal venous phase representative images are obtained from 45-90 seconds after the injection; 3) Delayed phase - representative images are obtained from 90 seconds up to 5 minutes after the injection. The typical appearance of classic HCC is similar to that of contrast-enhanced CT, with hypervascularity on arterial phase, and washout on portal venous and delayed phases (Figures 2-4). Iso/hypovascularity on arterial phase with slow or no washout on portal venous phase can be seen with variant HCC [23]. Recurrent or residual disease within previously ablated zone also demonstrates similar appearance (Figures 5 and 6). Percutaneous RFA Technique The electrode size is decided prior to the procedure based on the tumour size and the distance from the anticipated skin entry site to the tumour. Two RF ablation systems are utilized for treatment of liver tumours, LeVeen multitine electrode (Boston Scientific, Natick, MA) and the Cool-tip electrode (Covidien, Dublin, Ireland). The choice of electrode is based on operator preference and tumour location and is beyond the
Figure 2. Typical appearance of hepatocellular carcinoma on contrast-enhanced ultrasound. During arterial phase (left), the tumour demonstrates hypervascularity, with subsequent washout during portal venous and delayed phase imaging (right).
4
A. K. W. Chan et al. / Canadian Association of Radiologists Journal xx (2015) 1e8
Figure 3. Patient A: Triphasic liver computed tomography demonstrates a lesion in segment 4A of the liver, hypervascular on arterial phase (left image) with washout on both portal venous (middle image) and delayed phases (right image).
scope of this study. All ablations are performed by 1 of 3 interventional radiologists. Following adequate visualization with US, the skin entry site is cleaned and prepped and the site draped in a sterile fashion. Grounding pads provided by each of the vendors are placed on the upper thighs according to manufacturer recommendations. One percent Lidocaine is used to infiltrate the skin and under US guidance, the needle is advanced through the subcutaneous tissues to the liver capsule and further local anesthetic administered. Sedo-analgesia is then administered by a trained nurse intravenously and titrated throughout the procedure. A combination of Midazolam and Fentanyl are administered intravenously and continuous monitoring of blood pressure, pulse, oxygen saturation and respiratory rate performed by a trained nurse. A small incision is made using a no. 11 scalpel blade and under US guidance, the probe is inserted into the lesion and in the case of the LeVeen multitine needle, the tines deployed. The ablations are performed according to the manufacturer recommendations, in the case of the LeVeen multitine electrode, a specific algorithm is followed according to the size of the electrode chosen for the procedure and continued until satisfactory roll-off is achieved. In the case of the Cool-tip electrode, energy is delivered into the tumour for a total of 12 minutes and the procedure stopped regardless of whether roll-off was achieved.
If deemed necessary by the operator, further ablations are performed in order to achieve a minimum of 5 mm of healthy ablated tissue around the tumour. It is difficult to accurately assess the exact extent of the ablation zone and this is estimated based on the size of the tumour and the needle used for ablation. Following the ablation, the needle tract is ablated and the electrode removed. Results Pretreatment Imaging Prior to receiving their CEUS-guided RFA treatments, the patient’s pretreatment imaging (CT or MRI) was first reviewed by the 2 radiologists. Reader 1 reviewed 50 lesions, while reader 2 reviewed 47 lesions. A small number of patients (2 patients) had decreased renal function and therefore no iodinated or gadolinium based contrast agent was administered. Among the patients that received contrast, almost all lesions demonstrate the typical characteristics of hepatocellular carcinoma: hypervascular on arterial phase with subsequent washout on portal venous/delayed phase imaging. The characteristics of these lesions are summarized in Table 1. CEUS during Intervention
Figure 4. Patient A: Contrast-enhanced ultrasound (CEUS) performed 1 month later confirms a hypervascular lesion in segment 4A of the liver (Figure 3). Radiofrequency ablation was then performed under CEUS guidance.
Table 2 demonstrates the analysis of the lesions from patients who underwent CEUS-guided RFA. Reader 1 analyzed 50 lesions, 41 of these lesions were visualized with B-mode US and 37 lesions visualized with CEUS. Reader 2 analyzed 32 lesions, 27 of these lesions were visualized with B-mode US and 31 lesions visualized with CEUS. There is moderate interobserver agreement for both the B-mode and CEUS scoring, with kappa values of 0.455 and 0.543, respectively. Of the lesions visualized, the visualization score of lesions with CEUS is 2.3 0.7 for reader 1 and 2.4 0.8 for reader 2, while the visualization score of lesions with B-mode US is 1.7 0/7 for reader 1 and 1.8 0.8 for reader 2. The differences between the CEUS and B-mode are not statistically significant for either reader (P value of 0.734 for reader 1 and 0.950 for reader 2). Subjectively, however, reader 1 found CEUS was helpful compared to B-mode US alone in 70.3% of lesions, whereas reader 2 found CEUS helpful in 71.0% of lesions.
CEUS guided RFA treatment of HCC: a retrospective study / Canadian Association of Radiologists Journal xx (2015) 1e8
5
Figure 5. Patient A: Triphasic liver computed tomography at 6-month follow up demonstrates a hypervascular nodular focus within the segment 4A ablation zone, in keeping with recurrent disease (left image ¼ arterial phase; middle image ¼ portal venous phase; right image ¼ delayed phase).
Table 3 demonstrates the distribution of B-mode and CEUS visualization scores, while Table 4 compares the CEUS visualization scores to those of B-mode scores. Of the lesions not visualized by B-mode US, CEUS allows at least fair visualization (i.e. score 2) for 78% (7 of 9) and 80% (4 of 5) of the lesions by reader 1 and reader 2, respectively. Similarly, of the lesions poorly visualized by B-mode US (ie, score of 1), CEUS allows at least fair visualization (ie, score 2) for 50% (9 of 18) and 42% (5 of 12) of the lesions by reader 1 and reader 2, respectively. These data suggests that CEUS is more valuable for evaluation of lesions that are either not visualized or poorly visualized by B-mode US. Follow-Up Imaging Table 5 demonstrates the outcome of CEUS-guided RFA treatment on follow-up imaging. The vast majority of the follow-up was performed 1 month after the treatment, with a total number of 7 cases performed from 2-7 months after and 1 case performed 13 months after the treatment. In terms of imaging modality, almost all cases were performed with CT, except 1 case with MRI and 1 case with CEUS. Complete response (CR) was achieved in the vast majority of the lesions: 37 of 45 lesions (82.2% with 95% confidence interval of 11.2%) for reader 1, and 30 of 32 lesions (93.8% with 95% confidence interval of 8.4%) for reader 2, suggesting
CEUS-guided RFA is an efficacious treatment modality for patient with HCC at least in the short-term setting. Discussion RFA has become increasingly utilized in the treatment of HCC, due to it being minimally invasive, with lower morbidity rates compared to surgery [3]. Conventional B-mode US has been the standard modality used for guidance of probe placement. Unfortunately, visualization of the target tumour may not be adequate, and the identification of residual/recurrent disease is often difficult in the setting of previous ablation [4e6]. Since the introduction of secondgeneration microbubble US contrast agents, multiple studies have investigated the various roles of CEUS in management of patients with HCC. Several studies have confirmed the ability of CEUS to detect and characterize HCC [5e7,9e12,14e19,24e26]. Giorgio et al. [13] reported that the appearance of the lesions was concordant between CEUS and CT in 82.4% (61 of 74) of the cases, and there were no statistically significant differences in the visualization of the hypervascularity patterns of HCC between the 2 modalities. For detection of local recurrent disease, the morphologic patterns on CEUS are concordant with those on CT in all 17 cases of the study by Miyamoto et al. [5], and subsequently they successfully performed RFA in all of these recurrent cases. Table 1 Lesions characteristics on pretreatment imaging
Figure 6. Patient A: Contrast-enhanced ultrasound performed 1 week after the 6-month follow-up computed tomography (CT) (Figure 5) demonstrates the segment 4A recurrent disease, which closely resembles the appearance on CT (Figure 5).
Number of lesions Imaging modality CT MR Enhancement pattern No contrast administered 1 ¼ Hypervascular 2 ¼ Isovascular 3 ¼ Hypovascular Lesion type New lesion Residual disease (and pattern) 1 ¼ Peripheral 2 ¼ Central 3 ¼ Circumferential 4 ¼ Satellite
Reader 1
Reader 2
50
47
45 5
42 5
3 46 0 1
3 43 1 0
36 11 9 2 0 0
29 15 8 7 0 0
CT ¼ computed tomography; MR ¼ magnetic resonance.
6
A. K. W. Chan et al. / Canadian Association of Radiologists Journal xx (2015) 1e8
Table 2 Analysis of lesions from patients underwent CEUS-guided RFA
Number of lesions analyzed Number of lesions visualized with B-mode US B-mode score (mean SD) Number of lesions visualized with CEUS CEUS score (mean SD) Number of lesions where CEUS is helpful compared to B-mode alone Number of lesions where CEUS and pretreatment contrast-enhanced CT/MR are both available Number of lesions where CEUS is concordant with pretreatment CT/MR
Table 4 Comparison of B-mode and CEUS visualization scores
Reader 1
Reader 2
50 41 1.7 0.7 37 2.3 0.7 26 (70.3%)
32 27 1.8 0.8 31 2.4 0.8 22 (71.0%)
34
24
CEUS 30 (88.2%)
23 (95.9%)
Worse Equal Better Total
Not visualized
1
2
3
Total
N/A 2 7 9
7 2 9 18
5 7 5 17
3 3 N/A 6
15 14 21
Worse Equal Better Total
Not visualized
1
2
3
Total
N/A 1 4 5
6 1 5 12
3 2 4 9
2 4 N/A 6
11 8 13
CEUS ¼ contrast-enhanced ultrasound.
Not visualized
1
2
3
Total
superior to CEUS either, with Vilana et al. [25] reporting a sensitivity of only 20% (2 of 10 residual nodules detected). CEUS has proven to be much more accurate at evaluating treatment response at 1-month follow-up. Using contrastenhanced CT as the gold standard, the sensitivity/specificity of CEUS reported are: 91%/97% specificity (41 patients) by Vilana et al. [25], 92.3%/100% (100 patients) by Ricci et al [18], 83.3%/100% (110 nodules) by Salvaggio et al. [16], and 92.9%/98.8% by Bartolotta et al. [7]. Another 1-month follow-up study performed by Lu et al. [10] compared the diagnostic performance of both conventional B-mode US and CEUS to that of contrast-enhanced CT/MRI in 139 patients, in which the sensitivity/accuracy for B-mode US and CEUS are 78.2%/76.3% versus 97.0%/94.2%, respectively. With multiple studies demonstrating the ability of CEUS to detect and characterize HCC, CEUS can be used to guide RFA treatment and ultimately improve patient outcome. Minami et al. [6] retrospectively analyzed 66 patients with 108 hepatic malignancies (HCC and liver metastasis), in which technically successful ablation was achieved in 94% of patients in the first session. The unsuccessful cases were attributed to either deep tumour location, location behind costal bone, or insufficient enhancement. None of these patients demonstrated local tumour progression during follow-up imaging ranging from 1-12 months, although 5 patients from each of the HCC and metastasis groups demonstrated distant metastases in the liver. Maruyama et al. [4] conducted a prospective study consists of 55
2 0 4 3 9
7 2 3 6 18
2 3 7 5 17
1 1 1 3 6
12 6 15 17
Table 5 Outcome on short-term follow-up imaging
Not visualized
1
2
3
Total
1 0 1 3 5
6 1 2 3 12
1 2 2 4 9
0 1 1 4 6
8 4 6 14
Table 3 Distribution of B-mode and CEUS visualization scores B-mode
Not visualized 1 2 3 Total
CEUS
Reader 2
The use of CEUS immediately posttreatment is potentially of great benefit to the patients, since the detection of incomplete response would enable the delivery of additional treatment in the same session [7,26,27]. Dill-Macky et al. [26] reported a sensitivity of 88% and specificity of 40% in the ability of postprocedural CEUS to predict 2-week follow-up CT/MRI results in 21 lesions. The low specificity is likely related to the fact that it may be difficult to differentiate reactive hyperemia (either from tissue response or arteriovenous shunting) from residual tumour. It has been suggested that reactive hyperemia is usually uniform in thickness and surrounds the ablation zone in a rind-like fashion [27], whereas residual disease is often nodular or crescentic in shape. Evaluating treatment response at 24-hour posttreatment is also suboptimal, with Vilana et al. [25] reporting a sensitivity of 27.3% (3 of 11 residual nodules detected) when compared to contrast-enhanced CT at 1-month follow-up. This is due to obscuration of nodules from the presence of gas with strong posterior echoes. The use of contrast-enhanced CT at 24-hour posttreatment is not
CEUS
Reader 1
B-mode
CEUS ¼ contrast-enhanced ultrasound; CT ¼ computed tomography; MR ¼ magnetic resonance; RFA ¼ radiofrequency ablation; US ¼ ultrasound.
Reader 1
B-mode
Reader 1
B-mode Reader 2 CEUS
Not visualized 1 2 3 Total
CEUS ¼ contrast-enhanced ultrasound.
Complete response (CR) Partial response (PR) Stable disease (SD) Progressive disease (PD) Total
Reader 2
EASL
RECIST
EASL
RECIST
37 5 1 2 45
37 5 1 2 45
30 2 0 0 32
30 2 0 0 32
EASL ¼ European Association for the Study of RECIST ¼ Response Evaluation Criteria in Solid Tumors.
the
Liver;
CEUS guided RFA treatment of HCC: a retrospective study / Canadian Association of Radiologists Journal xx (2015) 1e8
hypervascular lesions (from 44 cirrhotic patients) diagnosed on contrast-enhanced CT but not visualized by conventional B-mode US. Among these 55 lesions, 53 were identified by CEUS, and 42 lesions were treated percutaneously with either RFA or ethanol ablation. All of these treated hypervascular lesions become hypovascular on posttreatment CT. As a result, they concluded that CEUS improves the detectability of lesions that are occult on conventional US and allows wider application of percutaneous treatment in cirrhotic patients. To our knowledge, this is the first study in North America to investigate the role of CEUS for guiding RFA treatment. The results from our study are largely concordant with those from the studies described above. First, we confirmed the ability of CEUS to increase conspicuity of lesions compared to B-mode US in general (mean visualization score of 2.3 0.7 vs 1.7 0.7 for reader 1 and 2.4 0.8 vs 1.8 0.8 for reader 2). However, the difference in visualization is more significant for lesions that are either not visualized or poorly visualized with B-mode US. For lesions not visualized by B-mode US, CEUS allows at least fair visualization (ie, score 2) for 78% (7 of 9) and 80% (4 of 5) of the lesions by reader 1 and reader 2, respectively. For lesions poorly visualized by B-mode US (ie, score of 1), CEUS allows at least fair visualization (ie, score 2) for 50% (9 of 18) and 42% (5 of 12) of the lesions by reader 1 and reader 2, respectively. The appearances on CEUS is largely concordant with those on contrast-enhanced CT or MRI (30 of 34 lesions ¼ 88% for reader 1; 23 of 24 lesions ¼ 96% for reader 2). The authors acknowledge limitations of this study: 1. This is small cohort of patients, and large studies would provide more robust data. Our data, however, support the short-term efficacy of CEUS as an adjunct to conventional B-mode US for guiding RFA treatments based on the EASL and RECIST criteria. 2. It is a retrospective study and a control group consisting of patients treated under the guidance of a modality that is currently used in most clinical settings (conventional B-mode US, contrast-enhanced CT), was not possible due to clinical constraints. As a result, although we were able to document the efficacy of CEUS guided RFA at 1-month follow-up in terms of the rate of residual or recurrent disease, superiority or equivalence compared to these modalities cannot be evaluated. To address this issue, we aim to perform a retrospective study of patients treated with RFA before CEUS has been implemented in our institution, thus allowing the analysis of outcomes in patients treated under the guidance of conventional B-mode US. One month is a short follow-up interval, and collection of longer-term outcome data (eg, 5 years) by evaluating subsequently follow-up imaging will be the next stage of the study. Other outcome variables such as mortality or disease-free survival could also be analyzed to further strengthen the study.
7
Conclusion CEUS increases tumour conspicuity and provides better characterization of hypervascular lesions that are either not seen or poorly seen on conventional B-mode US and provides real-time guidance of RFA with good short-term treatment responses. References [1] Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet 2012; 379:1245e55. [2] Bruix J, Llovet JM. Prognostic prediction and treatment strategy in hepatocellular carcinoma. Hepatology 2002;35:519e24. [3] Lencioni R, Cioni D, Crocetti L, et al. Early-Stage hepatocellular carcinoma in patients with cirrhosis: long-term results of percutaneous image-guided radiofrequency ablation. Radiology 2005;234:961e7. [4] Maruyama H, Takahashi M, Ishibashi H, et al. Ultrasound-guided treatments under low acoustic power contrast harmonic imaging for hepatocellular carcinomas undetected by B-mode ultrasonography. Liver Int 2009;29:708e14. [5] Miyamoto N, Hiramatsu K, Tsuchiya K, et al. Contrast-enhanced sonography-guided radiofrequency ablation for the local recurrence of previously treated hepatocellular carcinoma undetected by B-mode sonography. J Clin Ultrasound 2010;38:339e45. [6] Minami Y, Kudo M, Hatanaka K, et al. Radiofrequency ablation guided by contrast harmonic sonography using perfluorocarbon microbubbles (Sonazoid) for hepatic malignancies: an initial experience. Liver Int 2010;30:759e64. [7] Bartolotta TV, Taibbi A, Midiri M, et al. Hepatocellular cancer response to radiofrequency tumor ablation: contrast-enhanced ultrasound. Abdom Imaging 2007;33:501e11. [8] Shiozawa K, Watanabe M, Takayama R, et al. Evaluation of local recurrence after treatment for hepatocellular carcinoma by contrastenhanced ultrasonography using Sonazoid: comparison with dynamic computed tomography. J Clin Ultrasound 2010;38:182e9. [9] Koda M, Mandai M, Matono T, et al. Assessment of the ablated area after radiofrequency ablation by contrast-enhanced sonography; comparison with virtual sonography with magnetic navigation. Clin Imaging 2010;34:60e4. [10] Lu M-D, Yu X-L, Li A-H, et al. Comparison of contrast enhanced ultrasound and contrast enhanced CT or MRI in monitoring percutaneous thermal ablation procedure in patients with hepatocellular carcinoma: a multi-center study in China. Ultrasound Med Biol 2007;33:1736e49. [11] Sparchez Z, Radu P, Anton O, et al. Contrast enhanced ultrasound in assessing therapeutic response in ablative treatments of hepatocellular carcinoma. J Gastrointestin Liver Dis 2009;18:243e8. [12] Leen E, Kumar S, Khan SA, et al. Contrast-enhanced 3D ultrasound in the radiofrequency ablation of liver tumors. World J Gastroenterol 2009;15:289e99. [13] Giorgio A, Ferraioli G, Tarantino L, et al. Contrast-enhanced sonographic appearance of hepatocellular carcinoma in patients with cirrhosis: comparison with contrast-enhanced helical CT appearance. AJR Am J Roentgenol 2004;183:1319e26. [14] Kisaka Y, Hirooka M, Koizumi Y, et al. Contrast-enhanced sonography with abdominal virtual sonography in monitoring radiofrequency ablation of hepatocellular carcinoma. J Clin Ultrasound 2009;38:138e44. [15] Gallotti A, D’Onofrio M, Ruzzenente A, et al. Contrast-enhanced ultrasonography (CEUS) immediately after percutaneous ablation of hepatocellular carcinoma. Radiol Med 2009;114:1094e105. [16] Salvaggio G, Campisi A, Greco V, et al. Evaluation of posttreatment response of hepatocellular carcinoma: comparison of ultrasonography with second-generation ultrasound contrast agent and multidetector CT. Abdom Imaging 2009;35:447e53. [17] Numata K, Fukuda H, Ohto M, et al. Evaluation of the therapeutic efficacy of high-intensity focused ultrasound ablation of hepatocellular
8
[18]
[19]
[20]
[21]
[22]
A. K. W. Chan et al. / Canadian Association of Radiologists Journal xx (2015) 1e8 carcinoma by three-dimensional sonography with a perflubutane-based contrast agent. Eur J Radiol 2010;75:e67e75. Ricci P, Cantisani V, Drudi F, et al. Is contrast-enhanced US alternative to spiral CT in the assessment of treatment outcome of radiofrequency ablation in hepatocellular carcinoma? Ultraschall Med 2009;30:252e8. Luo W, Numata K, Morimoto M, et al. Role of Sonazoid-enhanced threedimensional ultrasonography in the evaluation of percutaneous radiofrequency ablation of hepatocellular carcinoma. Eur J Radiol 2010;75:91e7. Bruix J, Sherman M. American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma: an update. Hepatology 2011;53:1020e2. Riaz A, Memon K, Miller FH, et al. Role of the EASL, RECIST, and WHO response guidelines alone or in combination for hepatocellular carcinoma: radiologic-pathologic correlation. J Hepatol 2011;54:695e704. Forner A, Ayuso C, Varela M, et al. Evaluation of tumor response after locoregional therapies in hepatocellular carcinoma. Cancer 2009;115: 616e23.
[23] Jang H-J, Kim TK, Burns PN, et al. Enhancement patterns of hepatocellular carcinoma at contrast-enhanced US: comparison with histologic differentiation. Radiology 2007;244:898e906. [24] Luo W, Numata K, Morimoto M, et al. Clinical utility of contrastenhanced three-dimensional ultrasound imaging with Sonazoid: Findings on hepatocellular carcinoma lesions. Eur J Radiol 2009;72: 425e31. [25] 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;16:2454e62. [26] Dill-Macky MJ, Asch M, Burns P, et al. Radiofrequency ablation of hepatocellular carcinoma: predicting success using contrast-enhanced sonography. AJR Am J Roentgenol 2006;186:S287e95. [27] Solbiati L, Ierace T, Tonolini M, et al. Guidance and monitoring of radiofrequency liver tumor ablation with contrast-enhanced ultrasound. Eur J Radiol 2004;51(Suppl):S19e23.
