| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Radiology and Center for Imaging Science, Samsung Medical
Center, Sungkyunkwan University School of Medicine, 50, Ilwon-Dong,
Kangnam-Ku, Seoul 135-710, South Korea.
2 Present address: Department of Diagnostic Radiology, Tuen Mun Hospital, Tuen
Mun, New Territories, Hong Kong.
Received June 29, 2007;
accepted after revision December 14, 2007.
Address correspondence to H. Rhim
(forest{at}smc.samsung.co.kr).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Local tumor progression after radiofrequency ablation shows mostly the peripheral nodular type. The site of local tumor progression shows a higher concordance rate with insufficient ablative margin than contiguous vessel and subcapsular location.
Keywords: hepatic tumor • hepatocellular carcinoma • local tumor progression • metastases • radiofrequency ablation
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
With the increasing number of long-term survivors after radiofrequency ablation, it is predictable that the number of cases of local tumor progression will also increase accordingly. The reported rate of local tumor progression after various ablative techniques ranges from 2% to 60% [2, 5, 8-10]. Recognition of the morphologic pattern of local tumor progression would be helpful for early detection. The first purpose of our study was to observe the most common morphologic pattern of local tumor progression.
Various types of risk factors for local tumor progression have been proposed in different studies, with controversial results [11-18]. Several independent risk factors for local tumor progression proposed in the previous investigations related to the geometry of the ablation zone and liver anatomy including lack of peritumoral ablative margin, heat sink effect of contiguous vessels, and subcapsular location. However, to our best knowledge, there has been no study to assess how much the site of local tumor progression exactly correlated with insufficient ablative margin, contiguous large vessels, and subcapsular location, that is, concordance between local tumor progression at the susceptible sites with the risk factors. The second purpose of our study was to assess the morphologic pattern and exact site of local tumor progression with relation to the risk factors.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5
cm in maximum diameter, multinodular (up to three in number) tumor 3 cm
in maximum diameter each, the absence of extrahepatic metastases, prothrombin
time ratio > 50% and platelet count > 70,000 cells/mm3, and
Child-Pugh class A or B liver cirrhosis (in patients with HCC). For this retrospective study, we included 86 patients treated by radiofrequency ablation using an internally cooled electrode who showed local tumor progression during the follow-up. Sixty-two patients had 65 hepatocellular carcinomas and 15 patients had 17 hepatic metastases. For HCC, local tumor progression was defined as any enhancing lesion inside or abutting the radiofrequency ablation zone at the arterial phase with washout lesion at the delayed phase of the CT examination during follow-up. For metastases, enlargement or growth abutting the ablation zone would be labeled as local tumor progression. Tumor that was not contiguous with the ablative zone was considered new tumor rather than local tumor progression.
|
In the group with HCC, the tumor measured 0.9-4.8 cm in maximum diameter
(mean, 2.32 cm; median, 2.30 cm). Two tumors were smaller than or equal to 1
cm. Twenty-four tumors were larger than 1 cm and smaller than or equal to 2
cm. Twenty-seven tumors were larger than 2 cm and smaller than or equal to 3
cm. Eleven tumors were larger than 3 cm and smaller than or equal to 4 cm. One
tumor was larger than 4 cm. The diagnosis of HCC was confirmed by percutaneous
needle biopsies (n = 32), characteristic imaging findings and an
elevated tumor marker (
-fetoprotein > 400 g/L) (n = 21), or
typical imaging appearances by at least two radiologic findings (n =
12) [19].
In the group with metastases, the tumors measured 1.0-4.8 cm at their maximum diameters (mean, 2.70 cm; median, 2.50 cm). One tumor was smaller than or equal to 1 cm. Three tumors were larger than 1 cm and smaller than or equal to 2 cm. Eight tumors were larger than 2 cm and smaller than or equal to 3 cm. Two tumors were larger than 3 cm and smaller than or equal to 4 cm. Two tumors were larger than 4 cm. The diagnosis of metastasis was based on imaging findings correlated with known primary tumor. Both percutaneous radio frequency ablation (59 of the patients with HCC and 15 of the patients with meta stases) and intraoperative radiofrequency ablation (six of the patients with HCC and two of the patients with metastases) were included in the study. Follow-up time for patients with HCC ranged from 5.4 to 78.8 months (mean, 32.5 ± 17.6 months). Follow-up time for patients with metastases ranged from 6 to 35.6 months (mean, 22.3 ± 8.3 months).
Radiofrequency Ablation Procedures
The descriptions of the radiofrequency ablation procedures and data
assessment were based on the standardization of terms and reporting criteria
proposed by the Society of Interventional Radiology Technology Assessment
Committee and the International Working Group on Image-Guided Tumor Ablation
[20].
During the study period, all radiofrequency ablation procedures were performed with an internally cooled electrode (Cool-Tip Radiofrequency System, Valleylab). The radiofrequency system consists of a 480-kHz generator; a 15- or 20-cm-long, 17-gauge cooled-tip radiofrequency electrode with a 2- to 3-cm-long exposed metallic tip; and a dispersive pad applied to the patient's skin. Grounding was achieved by attaching a dispersive pad to each of the patient's thighs. We used 2- or 3-cm, single or clustered electrodes. All cases were performed under sonographic guidance (Acuson Sequoia 512, Siemens Medical Solutions). The procedures were performed by one of six experienced radiologists.
The strategy of radiofrequency ablation was to include a peripheral margin of 0.5 to 1.0 cm of normal hepatic parenchyma surrounding the tumor and the entire tumor. Tumors less than 2.5 cm were treated with one or two ablations; tumors greater than 2.5 cm were treated with multiple overlapping ablations or clustered electrodes. Electrode track ablation technique was also performed to minimize postprocedural bleeding and tumor seeding along the electrode track.
Imaging Protocol
All patients were followed up with serial unenhanced and triple-phase
contrast-enhanced CT of the liver. Follow-up CT was performed immediately
(< 24 hours), 1 month, then every 3 months after the ablation session. The
CT images were acquired using one of three helical scanners (HiSpeed or
LightSpeed QX/I, GE Healthcare, or Brilliance 40, Philips Medical Systems). A
total of 120 mL of nonionic contrast material (iopromide, 300 mg I/mL
[Ultravist 300, Bayer HealthCare]) was administered at a rate of 3 mL/s with
an automatic power injector. Images were obtained before and at 30, 60-70, and
180 seconds after IV contrast material injection. Scanning of all phases was
initiated at the dome of the right hemidiaphragm and continued caudally
through the entire liver.
Image Analysis
Two observers retrospectively evaluated the CT scans of the patients with
local tumor progression by consensus. Images were read on a PACS workstation
(Centricity RA1000, GE Healthcare). We assessed the largest diameter of the
index tumor, whether the tumor was abutting a vessel more than 3 mm in
diameter, and whether the tumor was subcapsular or away from the liver
capsule. A vessel was defined as contiguous if the distance between the index
tumor and the large vessel was less than 5 mm. The location was defined as
subcapsular if the distance between the index tumor and the hepatic capsule
was less than 5 mm.
At CT immediately after the ablation, we evaluated whether the peritumoral
ablative safety margin (ablative margin 
5 mm) surrounding the index tumor
was achieved or not. The hypoenhancing zone would be considered ablated
whereas the hyperemic zone surrounding the ablation zone would not be
considered ablated. If in doubt, we reassessed the margin at 1-month follow-up
CT. If there was insufficient ablative margin (< 5 mm), we recorded the
exact site of insufficient ablative margin. In addition, we assessed whether
there was a large (> 3 mm in diameter) vessel within 5 mm of the margin of
the index tumor. If there was a large vessel close to the index tumor, we
recorded the exact site of the large vessel with relation to the
radiofrequency ablation zone. The site of the liver capsule in relation to the
subcapsular tumor was also recorded.
The observers also evaluated the latent time to local tumor progression by calculating the duration between the date of radiofrequency ablation and the CT examination showing local tumor progression. Whenever there was a suspicious region that was finally confirmed to be local tumor progression on follow-up CT, the date of CT showing the suspicious region was counted as the date of local tumor progression. We evaluated the morphologic and locational pattern and the exact site of local tumor progression in relation to insufficient ablative margin, large contiguous vessel, and subcapsular location.
The morphologic pattern of local tumor progression is summarized in Figure 1. The site of local tumor progression was initially categorized as extrazonal or intrazonal relative to the ablation zone. If the local tumor progression protruded externally from the hypoenhancing ablation zone, it was defined as extrazonal. If the local tumor progression developed inside the ablation zone, it was defined as intrazonal. With modification of the classification system suggested by Chopra et al. [21] for extrazonal local tumor progression, local tumor progression would be further divided into three categories: peripheral nodular type, crescentic type, and circumferential type. For intrazonal local tumor progression, it would be further divided into three categories: peripheral nodular type, crescentic type, and gross enlargement type. When one or more new nodules developed at the margin of the ablation zone as focal enhancing nodules in HCC and hypoenhancing nodules in metastases, they would be labeled as peripheral nodular type. When a crescentic rim of new lesion developed around the margin of the ablation zone, it would be labeled as crescentic type. If the new lesion involved more than 50% of the circumference of the ablation zone, it would be labeled as circumferential type. Gross enlargement type would be denoted as gross total enlargement of the lesion relative to the ablation zone. In addition, for location related to the axis of the electrode, the pattern was divided into the distal part around the distal tip of the active electrode, middle part around the transverse diameter of the ablation zone, and proximal part around the proximal portion of the active electrode.
Data Analysis
The frequencies of the morphologic pattern and location of local tumor
progression were tabulated. The subsequent treatments for patients with local
tumor progression were documented. For com paring continuous variables
including latent time of local tumor progression between HCC and meta stases
groups, the Student's t test was used. For categoric analysis,
Fisher's exact test was used. All statistical analysis was performed using
SPSS software, version 10.0, for Microsoft Windows, and p values of
less than 0.05 were considered statistically significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
For hepatic metastases, the time to local tumor progression (Table 1) ranged from 96 to 926 days (mean, 275.7 days; median, 199.0 days). No local tumor progression occurred before 3 months, 14 (82.4%) occurred between 3 months and 1 year, and three (17.6%) occurred between 1 year and 2 years. Eight (47.1%) cases of local tumor progression were treated by radiofrequency ablation. Four (23.5%) cases of local tumor progression were treated by surgery. The remaining six (35.3%) cases of local tumor progression received no further invasive treatment. There was no significance difference between the local tumor progression time of HCC and metastases (p = 0.16).
|
|
|
|
|
|
The most common morphologic pattern of local tumor progression after radiofrequency ablation for metastases was extrazonal peripheral nodular in 13 cases (76.5%), followed by gross enlargement (Figs. 4A and 4B) in three (17.6%) and extrazonal crescentic (Fig. 5) in one (5.9%). Peripheral nodular type was significantly more common than the other types (p < 0.05).
|
|
|
|
|
In the group with metastases, 14 of 17 (82.4%) tumors with local tumor progression showed insufficient ablative margin after radiofrequency ablation. Thirteen of these 14 tumors with insufficient ablative margin showed local tumor progression at the exact site of insufficient ablative margin. Thus the concordance rate between the location of local tumor progression and insufficient ablative margin was 92.9% (13/14) (Table 3).
The overall concordance rate of local tumor progression with insufficient ablative margin was 86.8% (33/38).
Relationship of Local Tumor Progression Site to Contiguous Vessel
In the group with HCC, 19 (29.2%) of 65 tumors with local tumor progression
showed a contiguous vessel close to the index tumor. In other words, 70.8% of
local tumor progression was located at least 5 mm away from the index tumor.
The concordance rate between the location of local tumor progression and a
contiguous vessel was 57.9% (11/19) (Table
3), whereas 42.1% of local tumor progression did not develop at a
contiguous vessel site (Fig.
7). In the group with metastases, seven (41.2%) of 17 tumors with
local tumor progression showed a contiguous vessel close to the index tumor.
The concordance rate between the location of local tumor progression and a
contiguous vessel was 57.1% (4/7). The overall concordance rate of local tumor
progression with a contiguous vessel was 57.7% (15/26).
|
Relationship of Local Tumor Progression Site to Liver Capsule
In the group with HCC, 24 (36.9%) cases of local tumor progression occurred
at the subcapsular index tumor. The concordance rate between the locations of
local tumor progression with the liver capsule in the subcapsular tumor was
58.3% (14/24) (Fig. 7). In the
group with metastases, eight (47.1%) cases of local tumor progression occurred
at the subcapsular index tumor. The concordance rate between the locations of
local tumor progression with the liver capsule was 37.5% (3/8). The overall
concordance rate of local tumor progression with the liver capsule direction
was 53.1% (17/32).
The concordance rate of local tumor progression at the site of insufficient ablative margin was significantly higher than the concordance rate of local tumor progression adjacent to contiguous vessel and the liver capsule (p = 0.004). The percentage of local tumor progression at the site of insufficient ablative margin was also significantly higher than it was at the site of contiguous vessel and the liver capsule (p = 0.002).
Relationship of Local Tumor Progression Site to Electrode Axis
In the group with HCC, when considering relationship with the axes of
electrodes, most local tumor progression was around the transverse diameter of
the ablation zone (33/65, 50.8%), followed by the distal (19/65, 29.2%) and
proximal (13/65, 20.0%) sites. In the group with metastases, 16 (94.1%) cases
of local tumor progression developed around the transverse diameter of the
ablation zone. One (5.9%) local tumor progression developed at the distal site
of the ablation zone. Considering overall local tumor progression, 24.4%
developed distal to the electrode axis, 15.9% developed proximal to the
electrode axis, and 59.8% developed around the transverse diameter of the
ablation zone (Table 4). The
percentage of local tumor progression at the transverse diameter of the
electrode was significantly higher than at other sites (p <
0.05).
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Immediately after radiofrequency ablation, the radiofrequency ablation zone usually shows a sharply marginated low-attenuation nonenhancing ablation zone with a hyperemic rim at CT [10, 23, 24]. Acutely, the hypoenhancing region represents the necrotic zones. Any tumor cell in the hyperemic zone may still be viable. With time, the hyperemic zones should resolve and the ablated zone should involute. Comparison with the previous CT images is particularly important to detect subtle changes. Thus a baseline CT immediately after radiofrequency ablation is mandatory. Because the range of time to local tumor progression is variable, regular CT follow-up at 3-4 months seems reasonable to detect local tumor progression early and manage it properly. Recognizing the most common local tumor progression patterns can facilitate the earlier detection of local tumor progression in clinical practice and follow-up imaging.
The most common morphologic pattern of local tumor progression in our study was the extrazonal peripheral nodular type, irrespective of tumor type, which was similar to the results of previous studies [21]. Few (12.3%) local progressing tumors show intrazonal growth within the radiofrequency ablation zone. Gross enlargement pattern was only observed in patients with hepatic metastases. Because most metastatic nodules are hypoattenuating, it would be difficult to appreciate their growth in a hypoattenuating ablation zone. Such growth would usually be detected only when the nodule shows a gross increase in size from the ablation zone, rather than normal involution. Interpretation of images must also be correlated with tumor markers. In cases of an elevated tumor marker with the finding of negative local tumor progression, one should search for other intrahepatic or extrahepatic metastases. PET/CT may also be useful in such cases [25].
Many contributing factors to local tumor progression after radiofrequency
ablation have been suggested by various investigations
[7,
11,
12,
15]. They include tumor size,
tumor pathology, proximity to a major vessel, subcapsular location, approach
to ablation, ablative margin, and -fetoprotein. However, to our
knowledge, there has been no study to correlate the exact site of local tumor
progression with the geometric or anatomic factors. One of the aims of our
study was to assess how much these risk factors were precisely concordant with
the exact site of local tumor progression.
The current study showed a high concordance rate (86.8%) between the exact sites of local tumor progression and an insufficient ablative margin. Previous studies had shown the lack of a 5-mm ablative margin to be a risk factor for local tumor progression [12, 26]. However, whether the sites of local tumor progression were concordant with the insufficient ablative margin was not usually documented. With a high concordance rate of local tumor progression with insufficient ablative margin, our data further supports that sufficient ablative margin surrounding an index tumor is very important to minimize local tumor progression after radiofrequency ablation of hepatic tumors. Thus it is important to achieve a sufficient ablative margin and to evaluate the periablation area with insufficient ablative margin more carefully at follow-up CT.
Heat sink effect is defined as the decreased thermal effect of local ablation due to heat dissipated by blood flow [18, 27-30]. The presence of a vessel larger than 3 mm can cause incomplete perivascular tissue death after radiofrequency ablation. The shape of the ablation zone would be affected by the contiguous vessel and sufficient ablative margin may also be adversely affected. Several methods to decrease blood flow have been proposed, such as occlusion of the hepatic artery and Pringle maneuver [9, 31].
If the local tumor progression was attributed to the contiguous vessel, the site of the local tumor progression should be concordant with the vessel position relative to the ablation zone. Our study has shown that, overall, slightly more than half (57.7%) of the local tumor progression was concordant with the vessel location despite the presence of a contiguous vessel. This finding may indicate that when treatment effectiveness was determined, any remaining viable tumor cells may not necessarily survive along the vessel. In other words, about half of the local tumor progression developed at a site other than the sites of the contiguous vessel. It is important to document whether the local tumor progression site was concordant with the contiguous vessel before attributing local tumor progression to that contiguous vessel. Again, this has not been clearly documented in the other studies. Ignoring this factor may overestimate the actual number of local tumor progressions associated with the contiguous vessel.
Whether subcapsular location was a risk factor for local tumor progression, particularly with the percutaneous approach is controversial. Previous studies suggested subcapsular location as an independent risk factor for local tumor progression [14-16]. More recent studies suggested there was no difference in local tumor progression rates between subcapsular and central tumor after percutaneous radiofrequency ablation [17, 32]. Most studies, however, did not mention whether the exact sites of local tumor progression correlated with the liver capsule or not. In our study, if local tumor progression developed and the tumor was subcapsular, the overall concordance rate of local tumor progression at the liver capsule was 53.1%. About half of the local tumor progressions that developed in subcapsular tumors did not correlate with the liver capsule location, that is, the local tumor progression occurred toward the central part of liver. When treatment effectiveness is documented, if local tumor progression developed and is attributed to the liver capsule, the site of local tumor progression should also be at the subcapsular location. Ignoring whether the local tumor progression was concordant may result in overestimation of the association between local tumor progression and subcapsular location.
Two factors may contribute to an increased "reported local tumor progression rate" in subcapsular tumors. First, the subcapsular tumor may be incompletely ablated by operators due to fear of collateral injury to adjacent organs, diaphragm, and abdominal wall. Second, radiofrequency ablation of a subcapsular tumor can be painful to the patient and force the operator to decrease the duration of treatment and radiofrequency current used, thus affecting the completeness of treatment. If these factors played a role in local tumor progression for a subcapsular tumor, then the site of local tumor progression may not necessarily correlate with the liver capsule. A detailed documentation of the completeness of the ablation is important. If technical success is not achieved and if growth of residual visible tumor occurs, this growth should not be termed local tumor progression [20]. Technical success should be determined only when the tumor was treated according to protocol and was ablated completely [20]. Artificial ascites and pleural effusion may be helpful to decrease the risk of collateral thermal injury and reduce pain to the patient during treatment of a subcapsular tumor. This issue has been discussed in other studies [32-34].
There is no study that shows the exact site of local tumor progression related to the electrode axis. During radiofrequency ablation using an internally cooled electrode (Cool-Tip), the distal and proximal margins were usually the earlier sites of coagulation compared with the central part. The time lag between these regions may contribute to any difference in therapeutic efficacy and local tumor progression. In our study, concerning the site of local tumor progression related to the electrode axis, local tumor progression was more common around the transverse diameter of the ablation zone rather than the distal and proximal margins. The likelihood of local tumor progression around the transverse diameter of the radiofrequency ablation zone can be related to an insufficient ablative margin, heat sink effect from contiguous vessel, and intrinsic limitation of the straight-type electrode showing shorter transverse diameter of the radiofrequency ablation zone. Therefore, one should have a strategy for this issue in the planning phase because most local tumor progression in this area can be minimized by one or two overlapping ablations.
We assumed that local tumor progression should be uncommon at the distal or proximal margin along the electrode axis because we usually place the electrode tip at least 5 mm distal to the margin of the index tumor. However, the tip of the straight-type electrode could be retracted backward by the patient's respiration during the ablation, especially when a deep-seated index tumor was visible only in the deep inspiration state. Hence, an operator should monitor closely the location of the electrode during the ablation. In addition, if the proximal margin of the index tumor is close to the proximal margin of the active component of the electrode after placement of the electrode, additional ablation after retraction of the electrode about 1 cm is required to achieve sufficient ablative margin anterior to the index tumor.
The present study had several limitations. First, it was performed as a retrospective singlecenter study using single equipment without a control group. The pattern and exact site of local tumor progression may be different from our results when expandable or bipolar electrodes are used. Second, the assessment of ablative margin was sometimes difficult if the CT scans before and after radiofrequency ablation did not show identical anatomic orientation. In addition, in cases without coronal reformat images, the superior and inferior parts were particularly difficult to assess on axial images because of partial volume artifacts. Third, local tumor progression could not be pathologically proven. Finally, the variability of experiences and therapeutic planning of different interventional radiologists were not considered.
In conclusion, local tumor progression after radiofrequency ablation of hepatic tumors shows mostly the peripheral nodular type. There was high concordance between the site of local tumor progression and an insufficient ablation margin. This finding further supports the importance of a sufficient ablative margin for local tumor control. The concordance rate of risk factors with sites of local tumor progression is an important topic for a future research project.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
