Ablation manual for liver cancer

Abstract

Because of recent advances in energy device technology, ablation has become popular worldwide. It is less invasive and provides faster postoperative recovery compared to surgery, and therefore, it has come to be applied to a wide range of organs, such as liver, lung, kidney, thyroid, and bone/soft tissue tumors. In order to properly guide the needle to the target area, imaging support is necessary, and ultrasound, which has the advantages of high resolution and real-time capability, is the most frequently used modality. In other words, ablation can be said to be a therapeutic method that makes the most of the advantages of ultrasound. This article outlines the role of ultrasound in ablation for liver cancer and its specific usage.

Keywords: Ablation; Cancer; Hepatocellular carcinoma; Liver; Ultrasound.

Fig. 1

Planning. a HCC is found in two places in S3 of the liver. It is located ventrally in P3 and dorsally, respectively (arrows). It is depicted more clearly in P3 (arrowhead) when color Doppler is used. The dorsal lesion was to be ablated first, but it was determined that it was not appropriate as P3 was located on the puncture route when the puncture line was in the forward direction. b P3 could be circumvented from the puncture route by planning a puncture line in the opposite direction

Fig. 2

a Dedicated puncture probe. b Micro convex probe

Fig. 3

Anatomical structure that needs to be understood when creating artificial pleural effusion

Fig. 4

Structure of pneumoperitoneum needle

Fig. 5

Anatomical structure that needs to be understood when creating artificial ascites

Fig. 6

The flow of 5% glucose solution from the tip of the pneumoperitoneum needle (arrowhead) into the abdominal cavity is clearly depicted on color Doppler US (arrow)

Fig. 7

Image fusion system. The position and angle of the probe can be detected in real time in 3D space by means of a position detection system using magnetism. By specifying the shared sites (e.g., intrahepatic vessels) on CT volume data and the US image, an MPR image synchronized with a virtual US image could be demonstrated

Fig. 8

CT-US fusion imaging guidance. A hypoechoic nodule (round dotted line) with an indistinct tumor contour is shown on the US image (left side of screen) so that it corresponds with the densely stained HCC (arrow) on contrast-enhanced CT (right side of screen). This nodule is pierced with the RFA needle (arrowheads)

Fig. 9

Simulation using 3D Sim-Navigator and E-field. The expected ablation area (yellow) when the RFA needle is inserted along two virtual puncture lines (yellow and purple) is displayed. The simulation image can also be utilized for treatment navigation as a reference screen

Fig. 10

US-US overlay fusion. The right side of the screen is the baseline US image, where the HCC nodule is indicated in green. On the left side of the screen, the baseline US image is projected on a US image taken immediately after treatment using the overlay function (image overlay). A region differentiating the ablative hyperechoic area and the green tumor image can be detected as the margin

Fig. 11

a MWA (Emprint ablation system) was performed for HCC in S6, but the tip of the antenna is indistinct. b The antenna tip is clearly displayed when a needle navigation system (Cannon Medical Systems) that can display the needle tip is used

Fig. 12

Relationship between placement position of RFA puncture needle and extent of ablation. a In the case of an oval lesion, ablation with a sufficient margin can be achieved if it is punctured in the long axis direction. b In practice, it is not always easy to puncture an oval lesion with the long axis perfectly aligned. However, ablation with a sufficient margin can be achieved by inserting the needle along the long axis of the lesion, as shown in the figure. In some cases, one should consider setting the puncture line by inverting the probe left or right when puncturing the lesion with the needle. c If it is only possible to set the puncture line to be perpendicular to the long axis direction of the lesion, radiofrequency ablation is performed with multiple punctures as shown in the figure. d In the case of a lesion with a shape that is nearly a perfect circle, the margin should be made slightly larger, performing ablation in an oval manner. e If the needle is punctured so that it is perpendicular to the long axis of the oval lesion, the concern is that there is insufficient margin in the long axis of the lesion. f If the puncture is made far from the center of the lesion, the lesion may remain on the opposite side of the puncture site. g As opposed to the puncture in (f), it will be difficult to evaluate the site with suspected residual tumor due to post-ablation gas in the case of the puncture in (g). Therefore, moving the needle tip in the direction in (g) must be avoided during puncture. h Even if the puncture can be successfully performed in the longitudinal direction, there is a possibility of residual tumor deep in the lesion when ablation is performed at the front side of the lesion due to not being able to recognize the needle tip. Visualization of the deep part on US becomes difficult as gas appears with ablation of the superficial part, making evaluation of the part with suspected residual tumor challenging. Performing additional ablation of the deep part must be avoided as evaluation of the part with residual tumor and visualization of the needle tip both pose a high level of difficulty. i Ablation of the deep part of the lesion can leave residual tumor in the superficial part of the lesion. There is gas caused by ablation in the deep part, but the superficial part can be visualized, making it relatively easy to perform additional ablation. Therefore, performing a slightly deeper puncture makes it easier to evaluate the part with residual tumor and perform additional ablation. j In the case of a lesion that is a perfect circle, proceeding with treatment without performing ablation in an oval manner can leave residual tumor in the radial margin. In this case, it will be difficult to evaluate the margin on the deep side. Therefore, if the puncture line will likely be like that in (e), it will be important to envision performing ablation of the entire lesion by performing a puncture like that in (f), evaluating the presence or absence of residual tumor in the superficial part of the lesion with contrast-enhanced US, and then performing a puncture like that in (c). If the puncture line is expected to be like that in (j), it will be important to either make the longitudinal margin bigger or envision ensuring the margin by performing two punctures, as in (c). From the standpoint of increased difficulty of post-ablation evaluation, the puncture lines that need to be avoided most are those in (g) and (h), while the puncture lines in (f) and (i) can be corrected given that post-ablation evaluation is possible. However, the needle tip reaches the deep part in both (f) and (i); therefore, one must be attentive to complications such as vascular injury. Red indicates the tumor (lesion), and orange represents the extent of ablation

Fig. 13

RFA for hepatocellular carcinoma (right hepatic lobe anterior segment, tumor diameter 32 mm) and assessment of response immediately after treatment. a Check for the presence or absence of vessels on the puncture line via color Doppler US as a pre-treatment simulation. b Since the maximum diameter of the tumor is 32 mm and the longest diameter on the puncture line is about 32 mm, 3 cm was selected as the extent of ablation with the puncture needle. c An image of the puncture taken with intercostal scanning is shown. The tip (tip of current-carrying part, arrow) of the puncture needle and the end of the current-carrying part (arrowhead) are shown in the figure. The needle is located slightly to the right of center of the tumor when facing the screen. d The needle tip is seen on the right side of the tumor when the probe is placed horizontally to the patient's body and the tumor is observed in the horizontal direction. A portion of the right edge of the tumor cannot be seen due to being obscured by the ribs. e When the probe is rotated 90 degrees clockwise from the above-mentioned placement to observe the tumor in the vertical direction, it can be understood that the caudal side of the lesion will be primarily ablated given that the needle will be inserted in the tumor from the caudal side of the tumor and the needle tip is seen in the caudal side of the tumor. From the above, it can be understood three-dimensionally that the right caudal side of the target lesion will be ablated. f Gas caused by steam is observed at the center of the tumor during RFA treatment. g Second puncture. The second needle was inserted further to the right than the first needle. h In the case of intercostal scanning, the needle tip is seen on the right side of the tumor and more caudally than the first puncture when observing the tumor horizontally while scanning horizontally since the right side is punctured. A portion of the right edge of the tumor cannot be seen due to being obscured by the ribs. i The needle is inserted from the caudal side of the tumor and the needle tip is seen caudal to the center of the tumor when the tumor is observed in the vertical direction. The needle cannot be confirmed in the same cross section after the first needle as the puncture is made to the right of the first puncture site. j Steam gas is observed primarily on the right side of the tumor during RFA treatment. k The third puncture was made at the cranial midline of the tumor, and the fourth puncture was made slightly cranially and to the left side of the tumor. Gas caused by steam is observed in the entire tumor during the third ablation. l Treatment was ended after confirming the absence of contrast enhancement in an area larger than the original tumor in the arterial phase of low mechanical index contrast imaging performed to assess treatment response immediately after RFA. Steam gas is hyperechoic as it is immediately after treatment, and the image is slightly hyperechoic for that reason during contrast enhancement. Strict follow-up observation is necessary as the margin is less than 5 mm. Note that recurrence is not found 6 months after RFA. Arrowheads indicate the tumor border

Fig. 14

Right hepatic lobe posterior segment, recurrence of hepatocellular carcinoma (tumor diameter 10 mm) in vicinity of RFA. a On a fusion image of B-mode US and contrast-enhanced CT, a small site of recurrence in the vicinity of the RFA treatment site is seen in the arterial phase of contrast-enhanced CT (arrowhead), but the lesion itself cannot be detected due to being obscured by the lung on B-mode US. b The lesion was depicted as hypoechoic when it was observed in the post-vascular phase of low mechanical index contrast imaging after injection of artificial pleural effusion to ensure the visual field. c The same site was densely stained after a second intravenous injection of Sonazoid and recognized as the target hepatocellular carcinoma. d The hypoechoic lesion was targeted for puncture in the post-vascular phase of contrast-enhanced US. The markers for the tip and distal side of the puncture needle are depicted (arrows). e Steam gas was observed after performing RFA treatment at the same site. f Contrast enhancement is not seen at the RFA site on low mechanical index contrast imaging performed immediately after RFA. Based on a comparison of the arterial phase of contrast-enhanced CT before treatment (g) and the arterial phase of contrast-enhanced CT 1 month after treatment (h), it was determined that sufficient ablation was achieved. The arrowhead indicates the tumor border

Fig. 15

Hepatocellular carcinoma (left lobe lateral segment, tumor diameter 10 mm) that was difficult to detect on B-mode US. a On a fusion image of B-mode US and the arterial phase of EOB-MRI, the tumor cannot be detected on B-mode US at the site that corresponds to the site that is densely stained in the arterial phase of EOB-MRI. b The lesion was depicted as hypoechoic in the post-vascular phase of low mechanical index contrast imaging. c The same site was recognized as the target lesion after being densely stained with a second injection of Sonazoid. d The RFA needle (extent of ablation 15 mm) was inserted at the hypoechoic site in the post-vascular phase. Images d–g show transverse scanning after puncture. The puncture needle (arrow) is seen inside the tumor. e With respect to the target lesion, steam gas is found in an area more widespread than the hypoechoic site in the post-vascular phase. f The same site is not densely stained in the arterial phase of low mechanical index contrast imaging performed immediately after RFA, but the area around RFA was enhanced due to inflammation, etc. g On high mechanical index contrast imaging performed immediately after that, a wider area is observed as an unstained area as it is less affected by B-mode. Images h–k are images of longitudinal scanning before and after puncture. It is hypoechoic (h) in the post-vascular phase, and the puncture needle (arrow) is seen inside the tumor after puncture. j During ablation, widespread steam gas is seen in the target lesion. k The same site is not densely stained in the arterial phase on low mechanical index contrast imaging performed after RFA. Route ablation was performed when pulling out the needle as the portal vein was present on the puncture route. Sonazoid contrast enhancement is not found on the body surface side. l, m The arterial phase of EOB-MRI before RFA (l) and the arterial phase of contrast-enhanced CT 1 month after RFA (m) indicate that the tumor was properly treated. Arrowheads indicate the tumor border

Fig. 16

Hepatocellular carcinoma (left lobe medial segment, tumor diameter 10 mm) that was difficult to detect on B-mode US. a The lesion could not be detected on B-mode US with a linear probe. b Dense staining was seen in the arterial phase on low mechanical index contrast imaging. c It was depicted as hypoechoic in the post-vascular phase. d Puncture was performed in the arterial phase on low mechanical index contrast imaging on the day of RFA. The needle is seen in the densely staining part (arrow). e An insertion mark is seen on B-mode US after gas caused by ablation had disappeared. f An unstained area larger than the original lesion was seen in the arterial phase on low mechanical index contrast imaging performed after RFA. g No clear contrast findings are seen in the punctured area in the portal venous phase either. h Contrast enhancement is not seen up to the liver surface at the punctured area when observing the area while destroying bubbles in the portal venous phase on high mechanical index contrast imaging. With respect to the site stained by inflammation adjacent to the ablation zone in the arterial phase, high mechanical index contrast imaging was more effective for evaluation of the extent of ablation than low mechanical index contrast imaging as it is less affected by background B-mode US. In this case, evaluation was performed with simple MRI before RFA as a contrast agent could not be used with MRI due to the patient's chronic renal failure. In the medial segment of the left lobe, small hypointensity (i) was seen on the T1-weighted image, and small hyperintensity (j) was observed on the diffusion-weighted image. A change in the signal was seen in an area sufficiently larger than that of the original tumor on the T1-weighted image of simple MRI performed 1 month after RFA (k). This is thought to reflect the extent of necrosis. Arrowheads indicate the tumor border

Fig. 17

Assessment of response immediately after RFA using low mechanical index contrast imaging and high mechanical index contrast imaging. In the portal venous phase (a) of low mechanical index contrast imaging, it was difficult to evaluate contrast enhancement due to the background B-mode US and shielding by the ribs. After switching to high mechanical index contrast imaging (b), which is less affected by background B-mode US, it was possible to evaluate contrast enhancement at the ablation site by performing imaging while intermittently destroying the bubbles with frame rate 2. Contrast enhancement was not found at the RFA site in this case, indicating necrosis

Fig. 18

Hepatocellular carcinoma (right hepatic lobe posterior segment, tumor diameter 20 mm and 9 mm) that was difficult to detect on B-mode US. a On a fusion image of B-mode US with a convex probe and the arterial phase of contrast-enhanced CT, the tumor could not be detected at the site that corresponded to dense staining in the arterial phase of contrast-enhanced CT. b The lesion could also not be detected on B-mode US with a linear probe. c The lesion was depicted as hypoechoic in the post-vascular phase of low mechanical index contrast imaging. In addition to the liver surface, the lesion was seen in a slightly deep area in the arterial phase (d, g) and post-vascular phase (e, h) on low mechanical index contrast imaging on the day of RFA. RFA was performed for the respective target lesions in the post-vascular phase. Gas is seen in a more widespread area than the original lesion on B-mode US during RFA (f, i). j In the portal venous phase on high mechanical index contrast imaging performed after RFA, an area larger than the two lesions is observed as a non-contrast area. The axial section (k) and coronal section (m) in the arterial phase of contrast-enhanced CT before RFA, and the axial section (l) and coronal section (n) in the arterial phase of contrast-enhanced CT 1 month after RFA, indicate that the lesions were properly treated. However, we thought that caution was warranted as the axial section margin was less than 5 mm. Recurrence is not found 10 months after RFA. Arrowheads indicate the tumor border on the liver surface, and arrows show the tumor border in the slightly deep area

Fig. 18

Hepatocellular carcinoma (right hepatic lobe posterior segment, tumor diameter 20 mm and 9 mm) that was difficult to detect on B-mode US. a On a fusion image of B-mode US with a convex probe and the arterial phase of contrast-enhanced CT, the tumor could not be detected at the site that corresponded to dense staining in the arterial phase of contrast-enhanced CT. b The lesion could also not be detected on B-mode US with a linear probe. c The lesion was depicted as hypoechoic in the post-vascular phase of low mechanical index contrast imaging. In addition to the liver surface, the lesion was seen in a slightly deep area in the arterial phase (d, g) and post-vascular phase (e, h) on low mechanical index contrast imaging on the day of RFA. RFA was performed for the respective target lesions in the post-vascular phase. Gas is seen in a more widespread area than the original lesion on B-mode US during RFA (f, i). j In the portal venous phase on high mechanical index contrast imaging performed after RFA, an area larger than the two lesions is observed as a non-contrast area. The axial section (k) and coronal section (m) in the arterial phase of contrast-enhanced CT before RFA, and the axial section (l) and coronal section (n) in the arterial phase of contrast-enhanced CT 1 month after RFA, indicate that the lesions were properly treated. However, we thought that caution was warranted as the axial section margin was less than 5 mm. Recurrence is not found 10 months after RFA. Arrowheads indicate the tumor border on the liver surface, and arrows show the tumor border in the slightly deep area

Fig. 19

Hepatocellular carcinoma (right hepatic lobe posterior segment, tumor diameter 17 mm). a A circumscribed 17-mm hypointense lesion with an irregular contour is seen in the right lobe in the hepatobiliary phase on EOB-MRI. A circumscribed hypoechoic lesion is found on a fusion image with B-mode US. b RFA was performed at the same site. c Sonazoid contrast enhancement is not seen at the RFA site on a fusion image of the arterial phase of low mechanical index harmonic imaging immediately after RFA and the hepatobiliary phase of EOB-MRI before RFA. d By creating a fusion image of both using the enhanced portal vein and placing a GPS marker on the pre-RFA hepatobiliary phase of EOB-MRI, the GPS marker was depicted at the site corresponding to the tumor site in the portal venous phase of contrast-enhanced US immediately after RFA, and Sonazoid contrast enhancement was not found at the tumor site itself. On the other hand, it is difficult to evaluate the ablative margin due to the impact of gas and inflammation on the tumor body surface side. e Based on evaluation of the extent of necrosis on a fusion image comprised of the portal venous phase of high mechanical index contrast imaging and the hepatobiliary phase of EOB-MRI after RFA, it was determined that a sufficient margin was achieved. f A 17-mm hypointense lesion is seen in S7 in the hepatobiliary phase of EOB-MRI performed before treatment. g It was determined that the 17-mm lesion had been ablated in an area larger than that of the original lesion in the hepatobiliary phase of EOB-MRI performed after RFA treatment. h Local recurrence is not seen 3.5 years later. Arrowheads indicate the tumor border

Fig. 20

Hepatocellular carcinoma (right hepatic lobe anterior segment, tumor diameter 13 mm). a The lesion could not be detected on a fusion image of B-mode US and contrast-enhanced CT. b The lesion was depicted as hypoechoic in the post-vascular phase of low mechanical index contrast imaging. c The same site was densely stained after a second intravenous injection of Sonazoid and recognized as the target. d RFA was performed using a fusion image of the post-vascular phase of low mechanical index contrast imaging performed immediately after RFA (left screen) and the arterial phase of contrast-enhanced CT before RFA (right screen). The RFA needle pierces the lesion on the left screen. e Gas caused by RFA is seen. f Contrast enhancement is not seen at the RFA site immediately after RFA on the portal venous phase of high mechanical index contrast imaging. g By placing a GPS marker on the pre-RFA arterial phase of contrast-enhanced CT, the GPS marker was depicted at the site corresponding to the tumor site in the portal venous phase of contrast-enhanced US immediately after RFA, and absence of contrast enhancement at the RFA site was confirmed. h A 13-mm hypervascular lesion is seen in S8 in the arterial phase of contrast-enhanced CT performed before treatment. i It was determined that the 13-mm lesion had been ablated in an area larger than that of the original lesion in the arterial phase of contrast-enhanced CT performed after RFA. Arrowheads indicate the tumor border

Fig. 21

Hepatocellular carcinoma (right hepatic lobe posterior segment, tumor diameter 15 mm). a A 15-mm circumscribed, slightly hypoechoic lesion is seen in the posterior segment of the right lobe on B-mode US. b Dense staining in the arterial phase of low mechanical index harmonic imaging and washout in the portal venous phase 1.5 min after Sonazoid administration were seen (not shown). c It was hypoechoic in the post-vascular phase. d The deep part of the tumor was punctured with the extent of ablation set at 20 mm. e Monitoring was performed during ablation using a US-US fusion image that matched real-time B-mode US with B-mode spatial coordinates generated by acquiring 3D volume data immediately before that. An area larger than the original tumor was depicted as hyperechoic due to steam gas. f Shown here is a fusion image comprised of the portal venous phase of low mechanical index harmonic imaging immediately after RFA consisting of two punctures and the respective ablation and pre-RFA B-mode US. By placing a GPS marker on the pre-RFA B-mode US image, the GPS marker is depicted at the site corresponding to the tumor site in the portal venous phase of contrast-enhanced US immediately after RFA, as well. Given that contrast enhancement was not seen in an area larger than the target site, it was determined that appropriate treatment was achieved. The pathological diagnosis based on biopsy of the tumor performed immediately before RFA was poorly differentiated hepatocellular carcinoma. Based on a comparison of contrast-enhanced CT performed before and 1 month after treatment (g, h), it was determined that there was no residual tumor. i In the arterial phase of contrast-enhanced CT performed 4.5 years after RFA, shrinkage of the same site is seen, and recurrence is not found. Arrowheads indicate the tumor border

Fig. 22

a A blood flow signal heading in the direction of the abdominal wall was seen on color Doppler US when the RFA needle was removed after ablation. b It was observed for several minutes, but compression hemostasis was performed by pressing down on the area around the hemorrhagic spot with the tip of the pneumoperitoneum needle as the blood flow signal did not wane (arrow)

Fig. 23

The puncture line was observed with color Doppler US during puncture of hepatocellular carcinoma located behind a rib in S4. A blood flow signal is seen near the rib (arrow), which may be an intercostal artery