Radiofrequency ablation is a minimally invasive technique for treating primary or metastatic malignancies in humans.1,2 During RFA, an electrode is inserted into a lesion, and alternating electric current obtained from a generator is passed through the electrode's tip into the surrounding tissues. The electric current causes changes in the direction of ions, which creates ionic agitation and frictional heating around the electrode and results in tissue destruction by coagulative necrosis.3,4 Focal hyperthermia induces tumor destruction by a direct thermal effect in which a local increase in temperature causes immediate cell death through the interruption of cellular metabolism.5 Tumor destruction can continue after the cessation of the heat stimulus by an indirect thermal effect.6 In 1 study,4 a cool-tip RFA system created a larger ablation zone in less time and maintained the peak temperature for a longer period than did other RFA systems.
Distinct changes occur in the ablation zone and surrounding liver parenchyma after focal hyperthermia.5 Four zones of tissue changes are identifiable in the liver following RFA: the application center, central zone, transitional zone, and reference (normal tissue) zone. The application center is defined as the area proximal to the heat source.7,8 The central zone is the area immediately surrounding the application center and consists of necrotic tissue.3,5 The ablation zone refers to the application center and central zone combined. The transitional zone contains tissue that has subacute hemorrhage as a result of sublethal hyperthermic injury; therefore, cell death can progress further in the transitional zone.5 The reference zone describes normal tissue surrounding the transitional zone.
Assessment of therapeutic efficacy in the central ablation zone and transitional zone is essential to detect residual viable tissue or local recurrence after RFA, which would indicate the need for early retreatment.9,10 There have been many efforts to accurately evaluate RFA efficacy in the ablation zone and transitional zone of humans by use of various diagnostic imaging techniques, including CT, MRI, positron-emission tomography–CT, and ultrasonography.11 Contrast-enhanced imaging (eg, contrast-enhanced MRI, contrast-enhanced CT, and contrast-enhanced ultrasonography) can reveal blood perfusion of the target tissue in evaluations conducted after RFA.11 The disappearance of arterial enhancement in the ablation zone and adequate enhancement in the transitional zone indicate successful treatment.12 However, residual tumor tissue can be seen as an enhanced area along the margin of the ablation zone because the tumor tissue has an abundant blood supply as a result of neoangiogenesis.13,14 Thus, in human medicine, it is recommended that contrast-enhanced imaging be used to detect residual tumor > 1 month after RFA (ie, when the hyperemia reaction in the periablation zone has regressed).12,13
Use of CT perfusion can provide a quantitative and reproducible assessment of tissue perfusion by serial measurement of temporal changes in tissue density after a bolus injection of iodine contrast agent.14–16 This technique is based on the theory that tissue enhancement depends on the tissue iodine concentration, which indirectly reflects tissue vascularity and vascular physiology.9,17 In human medicine, CT perfusion has been used to grade tumors (it acts as an in vivo marker of neoangiogenesis) and to monitor tissue response to antiangiogenic drugs and transarterial chemoembolization in several neoplasms such as hepatocellular carcinoma.14,18 The potential role of CT perfusion for evaluation of the ablation and transitional zones after RFA in humans has been reported for studies9,19 in which remaining arterial blood flow was represented as an indicator of residual tumor.
Tissue stiffness as well as blood flow can be changed in the ablation zone after RFA. Protein denaturation occurs almost instantaneously in necrotic lesions when the temperature of the tissue exceeds 60°C, with an increase in tissue stiffness directly related to the thermal dose.20–23 Tissue stiffness can be assessed with ultrasonographic elastography, including SWE and strain elastography. Shear wave elastography can be used to estimate tissue stiffness qualitatively via a color-coded map and the results quantitatively recorded.24–26 In contrast to strain elastography, which uses manual compression for inducing strain, SWE has better reproducibility through a reduction in operator influence.27 In SWE, a transducer generates a longitudinal pulse, and the resulting tissue displacement is measured.28 The displacement value represents the off-axis propagation of the radiation force and is proportional to stiffness of the tissue.29 In humans, SWE is used on the target tissue during and immediately after RFA, and stiffness of the ablation zone and transitional zone is evaluated.30,31
In another study32 conducted by our research group, efficacy of RFA in livers of healthy Beagles was evaluated by use of contrast-enhanced ultrasonography, strain elastography, and contrast-enhanced CT, and we found that contrast-enhanced CT was the most reliable method for determining diameter of the ablation zone. Strain elastography could be used to define the margins of the ablation zone immediately after RFA. However, shadowing from microbubbles at that time can hinder ultrasonographic evaluation of the ablation area, and the subsequent peritoneal effusion formed can interfere sufficiently to induce strain.
The study reported here was conducted to assess the use of CT perfusion and SWE to evaluate acute changes in tissue perfusion and stiffness of the ablation zone in the liver of healthy dogs after RFA. We hypothesized that SWE can be used to assess changes in stiffness of the ablation zone immediately after RFA. We also hypothesized that the ablation zone, transitional zone, and normal hepatic parenchyma will have marked changes in stiffness and perfusion during early stages after RFA, and these changes (assessed by SWE and CT perfusion) will correspond with results for histologic examination.
Materials and Methods
Animals
Seven sexually intact male purpose-bred Beagles were used in the study. Dogs were 3 to 5 years old, and median body weight was 11.7 kg (range, 7.6 to 13 kg). All dogs were considered healthy on the basis of evaluation of systemic arterial blood pressure and results of a physical examination, CBC, serum biochemical analysis, and abdominal radiography and ultrasonography. Dogs were housed individually in cages and fed commercial dry food; water was available ad libitum. Protocols for the study were approved by the Institutional Animal Care and Use Committee of Chonnam National University, and animals were cared for in accordance with the Guidelines for Animal Experiments of Chonnam National University (CNU IACUC-YB-2017-54).
RFA
Food was withheld from dogs for 12 hours. Anesthesia was induced with an IM injection of a combination of medetomidinea and zolazepam-tiletamine.b Tramadol hydrochloridec (2 mg/kg) and cefazoline sodiumd (20 mg/kg) were also administered IV. Areas over the right and left hip joint of each dog were shaved, and grounding pads were placed on the shaved areas. Dogs were positioned in lateral recumbency, and peripheral regions of the liver were selected for RFA performed under ultrasound guidancee with a linear probef through a subcostal approach. Lesions were made by use of RFA performed on the basis of color Doppler ultrasonography (day 0). For 6 dogs, RFA was performed at 2 sites after the dogs were positioned in right and left lateral recumbency. For the remaining dog, RFA was performed at 3 sites after the dog was positioned in left and right lateral recumbency and dorsal recumbency.
The tip of a single cooled electrodeg (17 gauge and 20 cm long, with a 2-cm-long exposed portion) was placed in a selected site of the liver, and RFA was performed by use of an impedance-controlled deviceh with continuous delivery of chilled water by means of an RFA pumpi through the internal channels of the shaft. The chilled water maintained the temperature of the electrode tip below 12°C to prevent overheating of the electrode, which could have caused tissue carbonization. Ablation was performed only around the tip (the remainder of the electrode was insulated). Power initially was set at 30 W and then was increased at a rate of 10 W/min until RFA termination. The RFA was completed when impedance abruptly increased as a result of an increase of tissue resistance or 7 minutes had elapsed after starting RFA. The RFA pump was turned off, and the electrode was slowly retracted. After RFA, all dogs received cefazolin sodium (20 mg/kg, PO, q 12 h) and tramadol hydrochloride (2 mg/kg, PO, q 12 h).
Postablation procedures
Ablation lesions were evaluated with conventional B-mode ultrasonography and SWE immediately after RFA and with CT perfusion 30 minutes after RFA. Evaluations of the RFA lesions were repeated with all 3 modalities on day 4.
SWE
Stiffness of the ablation lesions was estimated immediately after RFA and on day 4 by use of SWEj with a linear-array transducer (12.0 to 17.0 MHz) by 2 investigators (JC and SP). For the SWE, high, intermediate, and low stiffness were color coded in red, green or yellow, and blue. The initial propagation map of SWE was widely created over the liver and extended from the ablation zone to normal liver parenchyma. Then, ROIs were placed over the ablation zone (red), portion surrounding the ablation zone (green or yellow), and normal parenchyma (blue) by selection of areas with regular and parallel lines in the propagation map to exclude artifacts or confounding factors (Figure 1).33 Quantitative data were obtained for each ROI.
Color-coded maps (A and C) and propagation maps (B and D) of SWE (squares) used to evaluate tissue stiffness in normal liver parenchyma (A and B) and an ablation lesion (C and D) created with RFA in the liver of a clinically normal dog. A—Notice that the normal liver parenchyma is uniformly blue. B—The lines in the normal liver parenchyma are parallel, and the interval between the lines is constant, which indicates the measurement value is reliable. C—Notice the wide range of colors in the ablation lesion attributable to variations in stiffness. D—Quantification of tissue stiffness is conducted with ROIs located in the ablation lesion in an area with parallel lines and uniform intervals to reduce inaccuracies from peritoneal effusion or air bubbles after RFA. The distorted and nonparallel lines in the propagation map make the data less reliable. Circles T1 and T2 and T7, T8, and T9 are ROIs. Scale on the side of each image is in centimeters.
Citation: American Journal of Veterinary Research 79, 11; 10.2460/ajvr.79.11.1140
CT perfusion
The RFA lesion in each dog was evaluated with CT perfusion on days 0 and 4. Anesthesia was induced with an IM injection of a combination of medetomidine (0.05 mg/kg) and zolazepam-tiletamine (1.25 mg/kg) and maintained with isoflurane.k Dogs were positioned in dorsal recumbency, and precontrast CT images were acquired by use of a 16 multi–detector row CT scannerl with the following settings: kV (peak), 120; mA, 130; slice thickness, 1 mm; and pitch, 1. Contrast-enhanced CT was then performed on the cranial portion of the abdomen to localize the scan level for CT perfusion after the injection of iodine contrast material. By use of results for the contrast-enhanced CT images, the scan field of view for CT perfusion was selected at an appropriate transverse level that included the ablation lesion. Then, an IV bolus injection of iodine contrast mediumm (880 mg of I/kg) was administered through a 20-gauge catheter into a cephalic vein at a rate of 4.5 mL/s by use of an automated injector.n A cine scan was performed 3 seconds after the start of the injection. Images for CT perfusion were acquired at the ablation lesions by use of the following settings: kV (peak), 110; mA, 80; slice thickness, 4.8 mm; slices per scan, 4; and scan time, 1 second. Almost all of the ablation zone was included in the CT images. Total scan time was 60 seconds, and a total of 240 images was acquired. A breath-hold technique (manual hyperventilation was used to induced apnea) was applied before each CT scan.
A qualitative map of liver perfusion was generated by use of the CT perfusion softwareo and displayed on the monitor in a color scale. Three round-to-oval ROIs were manually drawn over the ablation zone, periablation area, and normal parenchyma in each RFA lesion, as determined on the basis of the contrast-enhanced CT images. The ablation zone was defined as the unenhanced region, and the periablation area was defined as the contrast-enhanced rim around the ablation zone.32 Evaluation of tissue perfusion was based on the maximum slope model and performed by use of 1-compartment analysis as the mean slope of tissue enhancement divided by peak enhancement in the aorta.32 The time attenuation curve obtained from dynamic CT was divided into arterial and portal portions to calculate separate arterial and portal blood flow in the ablation lesion. Peak intensity in the spleen was used to mark the end of hepatic arterial enhancement.34 Several quantitative variables were determined, including blood flow (blood flow per mass of tissue per minute), blood volume (amount of blood contained in 100 g of tissue), and arterial liver perfusion (arterial fractional blood flow).
Histologic examination
After RFA lesions were evaluated on day 4, the anesthetized dogs were euthanized by IV injection of potassium chloride, and necropsy was performed. Shape, color, and stiffness of the RFA lesions were assessed grossly, and tissue samples of the lesions then were obtained. Tissue samples were fixed in neutral-buffered 10% formalin and embedded in paraffin for histologic examination. Then, 3-μm-thick sections of tissue, including ablation lesions and surrounding normal liver tissue, were stained with H&E stain. Slides were evaluated with light microscopy by 2 pathologists (K-OC and MJCA) to assess liver lesions.
Statistical analysis
Statistical analysis was performed with statistical software.p Data were reported as mean ± SD when applicable. A 1-way ANOVA was used to compare differences for SWE among the ablation zone, transitional zone, and normal parenchyma. Differences between times (day 0 and 4) were evaluated by use of the Mann-Whitney U test. A 1-way ANOVA was also used to assess whether there were differences in CT perfusion variables among the ablation zone, transitional zone, and normal parenchyma. A Mann-Whitney U test was used to compare CT perfusion variables between time points.
Results
A total of 15 lesions were created with RFA, and 14 lesions were evaluated 30 minutes and 4 days after ablation. One lesion was eliminated from further analysis because it was created in the interlobar area between 2 hepatic lobes. Liver RFA was safely performed in all dogs without any major adverse effects (eg, hemorrhage, thermal effects to adjacent organs, and infection). In 1 dog, peritoneal fat was ablated near the ablation lesion as a result of hyperthermic injury; fat necrosis was evident during necropsy, but there were no associated clinical signs. During RFA initiation, microbubbles developed within the ablation zone and generated shadowing artifacts in the distal aspect of the ablation area. Immediately after RFA, the margin of the ablation lesion was poorly defined on B-mode ultrasonography because of acoustic shadowing.
The SWE color-coded map displayed the ablation zone (hard tissue) in red, transitional zone in green or yellow, and surrounding normal parenchyma (soft tissue) in blue (Figure 2). Because shear waves could not propagate through extremely hard tissues, signal-void areas that were not color coded were visible. The margin of the ablation zone was clearly delineated from the transitional zone and normal parenchyma. Color-coded mapping of the 3 zones and clear delineation of the margin between each zone were similarly evident both 30 minutes and 4 days after RFA. Evaluation of the mean values of SWE for the ablation zone, transitional zone, and normal liver parenchyma revealed significant decreases from the inner to outer zones both 30 minutes and 4 days after RFA (Table 1). Mean values of SWE for the 3 zones did not change significantly between days 0 and 4.
Images for SWE of the liver 30 minutes (day 0; A and B) and 4 days (C and D) after RFA. On the color-coded maps (A and C), the ablation zone is red and represents the firmest tissue (green square). The transitional zone is yellow or light green, which indicates the middle range of tissue stiffness. Normal parenchyma is blue and represents the softest tissue. There are no changes on the color-coded map between days 0 and 4. For the 2-D mode (B), the ablation zone is hypoechoic (asterisk) and contains echogenic air bubbles (arrows) that cause acoustic shadowing (arrowheads) at 30 minutes after RFA. On day 4 (D), no hyperechoic air bubbles are visible around the ablation zone, and the ablation lesion is primarily hypoechoic (asterisk) and surrounded by a hyperechoic rim (arrows). Scale on the side of each image is in centimeters.
Citation: American Journal of Veterinary Research 79, 11; 10.2460/ajvr.79.11.1140
Stiffness values determined by use of SWE for the ablation zone, transitional zone, and normal liver parenchyma after RFA in 7 clinically normal dogs.
| Day* | Ablation zone | Transitional zone | Normal parenchyma |
|---|---|---|---|
| 0 | 118.6 ± 14.8a | 45.8 ± 9.8b | 12.6 ± 1.5c |
| 4 | 118.2 ± 12.1a | 46.7 ± 8.3b | 11.3 ± 2.2c |
Data represent mean ± SD kilopascals.
Day of RFA was day 0; samples on day 0 were obtained within 30 minutes after RFA.
Within a row, values with different superscript letters differ significantly (P < 0.05).
The color-coded map of CT perfusion displayed the ablation zone in dark purple (low blood perfusion), transitional zone in light green (large amount of blood perfusion), and normal liver parenchyma in dark purple to blue (Figure 3). The CT perfusion variables measured within the ablation zone, transitional zone, and surrounding normal parenchyma 30 minutes and 4 days after RFA were evaluated (Table 2). Quantitative perfusion variables, which included flow, volume, and arterial liver perfusion, were significantly higher in the transitional zone than in the ablation zone and normal parenchyma, and all variables of the ablation zone were significantly lower than for normal liver parenchyma on days 0 and 4. The CT perfusion variables did not change significantly between days 0 and 4, regardless of the tissue zone.
Images for the color-coded map of CT perfusion of an ablation lesion in the liver of a dog 30 minutes (day 0; A through D) and 4 days (E through H) after RFA. Images represent the maximum-intensity projection (HU; A and E), blood flow (mL/100 g of tissue/min; B and F), blood volume (mL/100 g of tissue; C and G), and arterial liver perfusion (mL/100 g of tissue/min; D and H). Evaluation of the central ablation zone (ROI 1) reveals a distinctive dark purple indicative of low blood perfusion, compared with results for the transitional zone (ROI 2) and normal liver parenchyma (ROI 3). The maximum-intensity projection for the transitional zone (ROI 2) has peripheral rim enhancement and is distinctive light green, which represents a large amount of blood perfusion. Normal liver parenchyma (ROI 3) is dark purple to blue.
Citation: American Journal of Veterinary Research 79, 11; 10.2460/ajvr.79.11.1140
Mean ± SD values for CT perfusion of the ablation zone, transitional zone, and normal liver parenchyma after RFA in 7 clinically normal dogs.
| Variable | Day* | Ablation zone | Transitional zone | Normal parenchyma |
|---|---|---|---|---|
| Flow (mL/100 g of tissue/min) | 0 | 1.5 ± 3.5a | 19.6 ± 9.3b | 7.6 ± 3.9c |
| 4 | 1.9 ± 3.2a | 35.7 ± 12.7b | 10.7 ± 7.3c | |
| Volume (mL/100 g of tissue) | 0 | 5.1 ± 8.2a | 84.6 ± 23.5b | 55.4 ± 21.8c |
| 4 | 9.4 ± 23.0a | 89.5 ± 33.3b | 58.3 ± 18.6c | |
| Arterial liver perfusion | 0 | 4.8 ± 8.2a | 31.5 ± 48.9b | 18.6 ± 38.1c |
| (mL/100 g of tissue/min) | 4 | 1.0 ± 0.8a | 27.2 ± 21.9b | 11.8 ± 18.5c |
Flow is blood flow per mass of tissue per minute. Volume is the amount of blood contained in 100 g of tissue. Arterial liver perfusion is the arterial fractional blood perfusion.
See Table 1 for remainder of key.
Gross and histologic examination 4 days after ablation revealed that ablated liver tissues had 3 distinct zones: ablation, transitional, and surrounding normal parenchymal (Figure 4). Gross examination revealed that the ablation lesion was a circular white region with a clear margin; palpation revealed the tissue had a firm texture. Histologic examination at low magnification revealed that the central ablation zone was severely pale and necrotic. At high magnification, each hepatocyte had evidence of coagulative necrosis, which was characterized by preservation of the basic outline of the cells together with pyknotic, karyorrhectic, or karyolytic nuclei. The transitional zone was macroscopically dark brown and microscopically had multiple areas of locally extensive congestion or hemorrhage (or both). In this zone, some lymphoid cells had evidence of neovas-cularization and fibroblast proliferation with active phagocytosis. There was also severe diffuse congestion, which was characterized by pale RBCs tightly packed within the sinusoids.
Photograph of the liver of a dog 4 days after RFA (A) and photomicrographs of tissue sections from the ablation lesion and surrounding tissues 4 days after RFA (B through E). A—A brownish area (bottom left) with a well-defined margin of surrounding darker brown tissue is visible. B—Three zones are visible during microscopic examination of the area indicated by the dotted region in panel A: inner ablation zone (a), middle transitional zone (t), and outer normal parenchymal zone (n). C—The central ablation zone is extremely pale, and there is necrosis with pyknosis. D—There is diffuse congestion with hemorrhage in the transitional zone, and the transitional zone is separated from the normal parenchyma by focal granulocytic infiltration of lymphoid cells and fibroblasts (asterisk). E—Notice the normal parenchyma in the outer zone. H&E stain; bar = 1 cm in panel A, 500 μm in panel B, and 50 μm in panels C, D, and E.
Citation: American Journal of Veterinary Research 79, 11; 10.2460/ajvr.79.11.1140
Discussion
Determination of RFA efficacy through adequate assessment of the ablation and transitional zones is crucial for successful ablation treatment. Acute physiologic changes of liver parenchyma were investigated in the study reported here, which focused on blood perfusion and tissue stiffness in dogs. Because RFA can cause coagulation necrosis of the target tissue, the ablation zone usually loses its blood flow completely; this is manifested as an unenhanced area on contrast-enhanced CT images.32 In contrast, the transitional zone usually has hyperemic changes attributable to congestion, which translates as an enhanced rim on contrast-enhanced CT images.32 However, this enhanced rim cannot be distinguished from residual tumor. Thus, further elucidation by use of quantitative data is needed.
In the present study, CT perfusion provided quantitative data (which included blood volume, blood flow, and arterial liver perfusion) for the ablation zone, transitional zone, and normal parenchyma. Examination of color-coded maps for CT perfusion revealed changes in tissue perfusion variables after RFA. Liver tissue receives blood via 2 distinct circulatory routes: approximately 75% to 80% is through the hepatic portal circulation, and the remaining 25% is through the hepatic artery from systemic circulation. A fundamental principle for CT perfusion is based on temporal changes in tissue attenuation.35 After IV administration of contrast media, changes in tissue attenuation reflect vascular physiology and tissue vascularity indirectly.35 Because of the complicated blood supply of the liver, measurement of a separate portal and arterial flow in the liver requires that the time attenuation curve generated from dynamic CT be divided into arterial and portal perfusion flow.34 Perfusion variables generated from CT perfusion represent the amount of blood entering the liver through the hepatic artery. Blood flow and volume within the ablation zone were significantly decreased 30 minutes after RFA and did not change on day 4. These changes of the perfusion variables agree with studies9,17 of humans in which blood volume and flow were almost totally absent in the ablation zone after RFA. For the study reported here, coagulation necrosis of the ablation zone caused by RFA was confirmed through histologic examination, which revealed severe diffuse cell death with necrosis and congestion that was characterized by pale RBCs tightly packed within the sinusoids. These findings represented coagulative necrosis and pyknosis in the ablation zone. Sudden heat stimuli can denature structural proteins as well as enzymes, so delayed proteolysis might have induced coagulative necrosis. Such protein denaturation would lead to a substantial decrease of perfusion variables in the ablation zone.
In the present study, arterial liver perfusion in the ablation zone was significantly less than in the transitional zone and normal parenchyma. Arterial liver perfusion can be used to predict residual tumor after RFA because this variable estimates arterial perfusion.18,36 The arterial blood supply to malignant hepatic tumors is increased; thus, arterial liver perfusion, total liver perfusion, and the hepatic perfusion index are increased. Results of CT perfusion can be used to predict the therapeutic response to chemoembolization in hepatocellular carcinoma of humans by assessing the amount of decrease in these 3 CT perfusion variables.16 Tumor response can be better predicted by use of the change in perfusion of malignant liver tumors, compared with assessments of tumor size.9 Although RFA and CT perfusion were not applied to malignant tumors in the present study, persistence of arterial perfusion after RFA would be indicative of incomplete ablation, the presence of collateral vessels, or recanalization of the tumor.37 In the study reported here, arterial liver perfusion was significantly lower in the ablation zone, which indicated complete ablation of arterial perfusion within the ablation zone.9,17
Thermal injury in the ablation zone also induced change in the stiffness of that zone. The ablation zone of the liver was shown in red on the SWE color-coded map and was clearly delineated from the transitional zone and normal parenchyma. The SWE color-coded map enabled us to localize the ablation zone within a wide range of liver tissue.
Quantitative SWE values were obtained within 30 minutes after RFA, even when gas bubbles were generated by the RFA procedure. Stiffness of the ablation zone was quantified via SWE, and the mean value was 10 to 11 times as high as that of the normal parenchyma. When RFA was performed in the liver of clinically normal rats, mean ± SD SWE values of the ablation zone, transitional zone, and normal parenchyma were 59.1 ± 21.9 kPa, 13.1 ± 1.5 kPa, and 4.3 ± 0.8 kPa, respectively.31 On the basis of the histologic findings and mean SWE values, the authors of that study31 suggested that the mean SWE values of the transitional zone can be used as a threshold for detecting irreversible cellular destruction and residual tumor. In addition, results of that study31 suggested that when there is an incomplete ablation zone, the transitional zone should be softer (< 13.1 kPa), compared with the value for a transitional zone surrounding a complete ablation zone.
In the present study, tissue stiffness did not differ on the basis of time point. The SWE values remained the same at both days 0 and 4. This result may have been attributable to instantaneous and irreversible cellular destruction in the ablation zone. Increased tissue stiffness is a result of cellular death involving tissue dehydration as well as protein denaturation. This result is consistent with results of another study.27
Use of CT perfusion revealed that blood volume, perfusion, and arterial liver perfusion in the transitional zone were significantly higher than in the ablation zone and normal parenchyma 30 minutes after RFA. These findings were supported by histopathologic observations (ie, congestion and hemorrhage in the transitional zone). Damage to the bile duct epithelium and vascular endothelium by thermal energy is believed to induce congestion in the transitional zone.38,39
In contrast, use of SWE revealed a value 2 to 3 times as high in the transitional zone as in the normal parenchyma. These changes may have been attributable to the congestion and edema elicited by thermal injury from the RFA procedure. Histopathologically, the transitional zone had mild focal granulocytic infiltration of lymphoid cells and fibroblasts; it also had marked hemosiderosis. These results can account for the substantial difference of perfusion and tissue stiffness of the transitional zone, compared with stiffness of the unablated normal parenchyma. Severe congestion with sinusoidal dilation can cause an increase in interstitial pressure,31 which may have accounted for the greater stiffness of the transitional zone, compared with stiffness of the normal parenchyma.
An increase in fibrous tissue, calcium and hemosiderin deposits, and polymorphonuclear histiocytes was observed in the transitional zone approximately 7 days after RFA in rabbit livers in a previous study40; however, there was no change in SWE of the transitional zone on day 4 after RFA in the study reported here. In another study,27 SWE performed on liver lesions of healthy pigs did not reveal a change in mean SWE values of each zone after RFA at various time points (0 to 60 minutes) after RFA. Thus, stiffness after RFA appears to be an immediate change that does not progress further.
For the present study, SWE values could be obtained within 30 minutes after RFA, even when gas bubbles were generated by the RFA procedure and despite the fact gas bubbles are the main obstacle for efficacy of ultrasonographic examination after RFA.32 In addition, tissue stiffness could be evaluated with SWE because a proper pulse could be created, even when a lesion was located in a deep area or an area with only an intercostal approach, and there was peritoneal effusion generated after RFA. In the study reported here, SWE of the superficial liver ablation was measured with the linear probe; however, SWE in deeper lesions can be measured with a microconvex ultrasound probe. The SWE propagation map helped us to place the ROIs over the RFA lesion. In SWE, a push pulse is applied to the tissue, and then shear waves are generated that extend perpendicular to the pulse. Reverberation or motion artifacts for the propagation of shear waves cause loss of direction and velocity of the shear wave; therefore, contour on the propagation map is distorted. In the present study, ROIs were placed (by use of guidance of a propagation map) over the area with parallel lines and consistent intervals to improve the reliability of SWE evaluation by eliminating effects of microbubbles generated during RFA.
Both CT perfusion and SWE were applied to dogs shortly after RFA. There were no complications related to imaging procedures in any dog in the present study. The color-coded maps for both procedures enabled the prediction of changes in perfusion and stiffness in the ablation zone and transitional zone, and quantitative data could be extracted for each area. Results effectively represented the physiologic and histologic changes of each area. In the study reported here, volume and size of the ablation lesion were not measured; however, SWE can provide morphological information (eg, volume and size) about lesions.27 A weakness of SWE is that the quantitative measurements can be affected by the machine and settings. On the other hand, CT perfusion directly reflects tumor microvasculature and can discriminate residual tumor from the transitional zone.9,17 When a clinician has trouble using CT to discriminate the transitional zone from the ablation zone after RFA, CT perfusion can provide a color-coded map for lesion distribution.9 However, there are limitations of CT perfusion, which include motion artifacts and limited coverage of the entire lesion. Repetitive examination of ablation lesions by use of CT perfusion is not recommended because of the high radiation dose.
Perfusion and stiffness were monitored on day 4 after RFA in the present study; however, congestion and edema in the transitional zone can last for > 1 month after ablation.9 Thus, higher values for perfusion variables such as blood volume, flow, and arterial liver perfusion can persist (for at least 1 month) in the transitional zone, but they should gradually decrease slowly over time. These results can differ in the presence of residual tumor, tumor recurrence, or cirrhotic parenchyma at ≥ 1 month after RFA.9,17 On the basis of results of other studies,9,17 we propose that there are differences in values of perfusion variables between the transitional zone and residual tumor at 1 month after RFA; perfusion variables of a transitional zone will gradually decrease, but those variables will remain high when there is residual tumor. As a result, serial changes of tissue perfusion and stiffness may provide data that are useful for predicting treatment response after RFA.
The present study had some limitations. First, it was not possible to ensure that the histologic sample was obtained from the exact same site where imaging data were measured by use of SWE and CT perfusion. However, because the entire RFA area was evaluated histologically and the ROIs were as large as possible for SWE and CT perfusion, sample site should not have had a major impact on results. Second, it was unclear whether these measurement techniques would be capable of distinguishing residual tumor from the edge of the transitional zone. However, on the basis of results of previous studies,17,27 it was expected that these 2 imaging methods could be used to distinguish residual tumor from the transitional zone earlier after RFA than other evaluation methods. Third, RFA was performed in normally vascularized liver, and changes in tissue stiffness and perfusion could differ from those in pathological conditions, particularly with altered internal vascularity. Therefore, changes in stiffness and perfusion of hepatic tumors should be evaluated by use of SWE and CT perfusion and compared with changes observed in normal tissues of the present study. Fourth, SWE of the ablation lesion was measured by use of a propagation map over the area with parallel lines and consistent intervals. However, bias could have been involved in the placement of ROIs over the ablation area, compared with placing ROIs that included all tissue of the 3 zones.
For the study reported here, acute changes in liver tissue after RFA were effectively evaluated by use of CT perfusion and SWE. The ablation zone, transitional zone, and normal hepatic parenchyma had marked changes in stiffness and perfusion as soon as 30 minutes after RFA, and these changes corresponded to results for the histologic examination. Coagulation necrosis induced a complete loss of blood perfusion in the ablation zone 30 minutes after RFA, and blood volume, blood flow, and arterial liver perfusion were significantly lower, compared with results for the transitional zone and normal parenchyma. Protein degeneration attributable to thermal injury increased tissue stiffness that could be assessed with SWE, even in the presence of gas bubbles generated by RFA. Congestion and edema in the transitional zone induced a significant increase in blood perfusion, as determined with CT perfusion. Tissue in the transitional zone was softer than in the ablation zone; however, it was markedly harder than the normal parenchyma. The distinct delineation of the ablation zone from the other zones on the color map as well as the SWE values made it possible to reliably determine the margin of each zone. Thus, SWE could be used to assess the ablation and transitional zones 30 minutes after RFA in all dogs, and ROIs for SWE could be placed reliably by use of the propagation map. Quantification of blood perfusion and stiffness of the ablation and transitional zones can provide information for use in predicting efficacy of RFA by determining the ablation margin and presence or absence of residual or recurrent tumor.
Acknowledgments
Supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (NRF-2018R1A2B6006775).
ABBREVIATIONS
| RFA | Radiofrequency ablation |
| ROI | Region of interest |
| SWE | Shear wave elastography |
Footnotes
Domitor (0.05 mg/kg), Orion Corp, Espoo, Finland.
Zoletil (1.25 mg/kg), Virbac, Carros, France.
Tamadol, Dongkwang Pharm, Seoul, Republic of Korea.
Cefozol, Hankook Korus Pharm, Seoul, Republic of Korea.
ProSound α7, Hitachi-Aloka, Tokyo, Japan.
UST-5412 (5 to 13 MHz), Hitachi-Aloka, Tokyo, Japan.
Cool-Tip RF electrode, Covidien, Burlington, Mass.
Cool-Tip RF generator, Radionics/Valleylab/Covidien, Burlington, Mass.
Cool-Tip RF ablation pump, Covidien, Burlington, Mass.
Aplio 500, Toshiba Medical System, Tochigi, Japan.
Terrell, Piramal Critical Care, Bethlehem, Pa.
Somatom Emotion, Siemens Medical Systems, Berlin, Germany.
Omnihexol 300, Korea United Pharm Co, Seoul, Republic of Korea.
Medrad Vistron C-T injector system, Medrad Inc, Warren-dale, Pa.
Perfusion CT software, Siemens, Erlangen, Germany.
SPSS Statistics, version 20, IBM Corp, Armonk, NY.
References
1. Ahmed M, Solbiati L, Brace CL, et al. Image-guided tumor ablation: standardization of terminology and reporting criteria—a 10-year update. J Vasc Interv Radiol 2014;25:1691–1705, 1705.e1–1705.e4.
2. Claudon M, Dietrich C, Choi B, et al. Guidelines and good clinical practice recommendations for contrast enhanced ultrasound (CEUS) in the liver—update 2012: a WFUMB-EFSUMB initiative in cooperation with representatives of AFSUMB, AIUM, ASUM, FLAUS and ICUS. Ultraschall Med 2013;34:11–29.
3. Chu KF, Dupuy DE. Thermal ablation of tumours: biological mechanisms and advances in therapy. Nat Rev Cancer 2014;14:199–208.
4. Quaranta V, Manenti G, Bolacchi F, et al. FEM analysis of RF breast ablation: multiprobe versus cool-tip electrode. Anticancer Res 2007;27:775–784.
5. Vanagas T, Gulbinas A, Sadauskiene I, et al. Apoptosis is activated in an early period after radiofrequency ablation of liver tissue. Hepatogastroenterology 2009;56:1095–1099.
6. Nikfarjam M, Muralidharan V, Christophi C. Mechanisms of focal heat destruction of liver tumors. J Surg Res 2005;127:208–223.
7. Germer C, Isbert CM, Albrecht D, et al. Laser-induced thermotherapy for the treatment of liver metastasis. Correlation of gadolinium-DTPA-enhanced MRI with histomorphologic findings to determine criteria for follow-up monitoring. Surg Endosc 1998;12:1317–1325.
8. Ohno T, Kawano K, Sasaki A, et al. Expansion of an ablated site and induction of apoptosis after microwave coagulation therapy in rat liver. J Hepatobiliary Pancreat Surg 2001;8:360–366.
9. Ippolito D, Bonaffini PA, Capraro C, et al. Viable residual tumor tissue after radiofrequency ablation treatment in hepatocellular carcinoma: evaluation with CT perfusion. Abdom Imaging 2013;38:502–510.
10. Tsui PH, Wang CY, Zhou Z, et al. Monitoring radiofrequency ablation using ultrasound envelope statistics and shear wave elastography in the periablation period: an in vitro feasibility study. PLoS One 2016;11:e0162488.
11. Nikfarjam M, Malcontenti-Wilson C, Christophi C, et al. Focal hyperthermia produces progressive tumor necrosis independent of the initial thermal effects. J Gastrointest Surg 2005;9:410–417.
12. Guan YS, Sun L, Zhou XP, et al. Hepatocellular carcinoma treated with interventional procedures CT and MRI follow-up. World J Gastroenterol 2004;10:3543–3548.
13. Park MH, Rhim H, Kim YS, et al. Spectrum of CT findings after radio-frequency ablation of hepatic tumors. Radiographics 2008;28:379–392.
14. Sahani DV, Holalkere NS, Mueller PR, et al. Advanced hepatocellular carcinoma: CT perfusion of liver and tumor tissue— initial experience. Radiology 2007;243:736–743.
15. Miles KA, Griffiths MR. Perfusion CT: a worthwhile enhancement? Br J Radiol 2003;76:220–231.
16. Oğul H, Kantarcı M, Genç B, et al. Perfusion CT imaging of the liver: review of clinical applications. Diagn Interv Radiol 2014;20:379–389.
17. Ippolito D, Fior D, Bonaffini PA, et al. Quantitative evaluation of CT perfusion map as indicator of tumor response to transarterial chemoembolization and radiofrequency ablation in HCC patients. Eur J Radiol 2014;83:1665–1671.
18. Ippolito D, Sandro S, Massimo P, et al. Perfusion computed tomographic assessment of early hepatocellular carcinoma in cirrhotic liver disease: initial observations. J Comput Assist Tomogr 2008;32:855–858.
19. Meijerink MR, van Waesberghe JH, van der Weide L, et al. Early detection of local RFA site recurrence using total liver volume perfusion CT. Acad Radiol 2009;16:1215–1222.
20. Rempp H, Hoffmann R, Roland J, et al. Threshold-based prediction of the coagulation zone in sequential temperature mapping in MR-guided radiofrequency ablation of liver tumours. Eur Radiol 2012;22:1091–1100.
21. Zhou Z, Wu W, Wu S, et al. A survey of ultrasound elastography approaches to percutaneous ablation monitoring. Proc Inst Mech Eng H 2014;228:1069–1082.
22. Brosses ES, Pernot M, Tanter M. The link between tissue elasticity and thermal dose in vivo. Phys Med Biol 2011;56:7755–7765.
23. Mariani A, Kwiecinski W, Pernot M, et al. Real time shear wave elastography monitoring of thermal ablation: in vivo evaluation in pig livers. J Surg Res 2014;188:37–43.
24. Ianculescu V, Ciolovan LM, Dunant A, et al. Added value of Virtual Touch IQ shear wave elastography in the ultrasound assessment of breast lesions. Eur J Radiol 2014;83:773–777.
25. Golatta M, Schweizer-Martin M, Harcos A, et al. Normal breast tissue stiffness measured by a new ultrasound technique: virtual touch tissue imaging quantification (VTIQ). Eur J Radiol 2013;82:e676–e679.
26. Golatta M, Schweizer-Martin M, Harcos A, et al. Evaluation of virtual touch tissue imaging quantification, a new shear wave velocity imaging method, for breast lesion assessment by ultrasound. Biomed Res Int 2014;2014:960262.
27. Bo XW, Li XL, Xu HX, et al. 2D shear wave ultrasound elastography (SWE) evaluation of ablation zone following radio-frequency ablation of liver lesions: is it more accurate? Br J Radiol 2016;89:20150852.
28. Eyerly SA, Vejdani-Jahromi M, Dumont DM, et al. The evolution of tissue stiffness at radiofrequency ablation sites during lesion formation and in the peri-ablation period. J Cardiovasc Electrophysiol 2015;26:1009–1018.
29. Sarvazyan AP, Rudenko O, Swanson SD, et al. Shear wave elasticity imaging: a new ultrasonic technology of medical diagnostics. Ultrasound Med Biol 1998;24:1419–1435.
30. Arnal B, Pernot M, Tanter M. Monitoring of thermal therapy based on shear modulus changes: II. Shear wave imaging of thermal lesions. IEEE Trans Ultrason Ferroelectr Freq Control 2011;58:1603–1611.
31. Sugimoto K, Oshiro H, Ogawa S, et al. Radiologic-pathologic correlation of three-dimensional shear-wave elastographic findings in assessing the liver ablation volume after radio-frequency ablation. World J Gastroenterol 2014;20:11850–11855.
32. Moon S, Park S, Lee SK, et al. Comparison of elastography, contrast-enhanced ultrasonography, and computed tomography for assessment of lesion margin after radiofrequency ablation in livers of healthy dogs. Am J Vet Res 2017;78:295–304.
33. Lee ES, Lee JB, Park HR, et al. Shear wave liver elastography with a propagation map: diagnostic performance and inter-observer correlation for hepatic fibrosis in chronic hepatitis. Ultrasound Med Biol 2017;43:1355–1363.
34. Zwingenberger AL, Shofer FS. Dynamic computed tomographic quantitation of hepatic perfusion in dogs with and without portal vascular anomalies. Am J Vet Res 2007;68:970–974.
35. García-Figueiras R, Goh VJ, Padhani AR, et al. CT perfusion in oncologic imaging: a useful tool? AJR Am J Roentgenol 2013;200:8–19.
36. Ippolito D, Bonaffini PA, Ratti L, et al. Hepatocellular carcinoma treated with transarterial chemoembolization: dynamic perfusion-CT in the assessment of residual tumor. World J Gastroenterol 2010;16:5993–6000.
37. Yang L, Zhang XM, Zhou XP, et al. Correlation between tumor perfusion and lipiodol deposition in hepatocellular carcinoma after transarterial chemoembolization. J Vasc Interv Radiol 2010;21:1841–1846.
38. Swift B, Strickland A, West K, et al. The histological features of microwave coagulation therapy: an assessment of a new applicator design. Int J Exp Pathol 2003;84:17–30.
39. Kotoh K, Morizono S, Kohjima M, et al. Evaluation of liver parenchymal pressure and portal endothelium damage during radio frequency ablation in an in vivo porcine model. Liver Int 2005;25:1217–1223.
40. Kato T, Suto Y, Hamazoe R, et al. Effects of microwave tissue coagulation on the livers of normal rabbits: a comparison of findings of image analysis and histopathological examination. Br J Radiol 1996;69:515–521.
