DOI:10.2214/AJR.04.1750
AJR 2006; 186:S284-S286
© American Roentgen Ray Society
CT Artifact Introduced by Radiofrequency Ablation
Darren D. Brennan1,
Liat Appelbaum1,
Vassilios Raptopolous1,
Jonathan B. Kruskal1 and
S. Nahum Goldberg1
1 All authors: Department of Abdominal Imaging and Intervention, Beth Israel
Deaconess Medical Center, One Deaconess Rd., Boston, MA 02215.
Received November 11, 2004;
accepted after revision February 21, 2005.
Address correspondence to D. D. Brennan
(dbrennan{at}caregroup.harvard.edu).
Keywords: CT fluoroscopy • interventional radiology • physics • radiofrequency ablation
Introduction
Imaging-guided percutaneous radiofrequency ablation of focal neoplastic
disease is an increasingly popular method of local tumor control and is being
applied to liver, lung, kidney, adrenal, and bone tumors
[1]. CT fluoroscopy can be used
for imaging guidance when percutaneous radiofrequency ablation is applied. CT
fluoroscopy can enable accurate tumor localization, intraprocedural
monitoring, and procedure control. In this article, we describe a case of
severe degradation of CT fluoroscopic images apparently caused by active
radiofrequency application. This finding could have implications on the choice
of imaging equipment and technique for guiding radiofrequency ablation.
Case Report
The reported case concerns a 71-year-old man with a history of idiopathic
cirrhosis since 1996. Sequential follow-up showed a rise in
-fetoprotein level. He underwent CT, which showed four discrete
hypervascular tumors. At biopsy, they were confirmed to be hepatocellular
carcinoma. He was subsequently referred for transarterial chemoembolization,
followed by successful radiofrequency ablation in 2003. A 4.2-cm tumor then
developed in segment IV, and he again was referred for radiofrequency
ablation.
The procedure was performed using an internally cooled electrode (Cool Tip,
Valleylab) and followed the manufacturer's recommended pulsing algorithm of
radiofrequency deposition [2].
A 17-gauge cluster electrode was advanced under CT fluoroscopic guidance into
the lesion. Guidance was provided by an Aquilon 64-MDCT volume scanner
(Toshiba America Medical Systems) according to the manufacturer's
recommendations. CT fluoroscopic image parameters were 120 kVp, 40 mAs, and
three slices 8-mm thick. After electrode placement had been confirmed
(Fig. 1A), ramped
radiofrequency energy was applied at 2,000 mA for a total of 14 min, using an
impedance-controlled algorithm and a 200-W, 480-kHz output generator (model
CC-1, Valleylab). CT fluoroscopy was used to monitor therapy while the
radiofrequency current was ramped to 1,200, 1,600, and 2,000 mA. During
successive imaging at these three currents, we saw severe and equivalent image
degradation (Figs. 1B and
1C). Later in the application,
when the radiofrequency was pulsed, we showed repeatedly that image
degradation ceased when the current was pulsed off
(Fig. 1D) but recurred when the
current was pulsed on. In addition, we showed that an increase from 40 to 250
mAs during CT fluoroscopic acquisition reduced, but did not eliminate, this
degradation (Fig. 1E). The
procedure was concluded without adverse events and with adequate ablation of
tumor margins, and the patient remains well 3 months after the procedure.
View larger version (132K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —CT fluoroscopy images of electrode placement in 71-year-old man with
hepatocellular carcinoma. Image obtained with 120 kVp, 40 mAs, and 8-mm slice
thickness during radiofrequency ablation using a 480-kHz generator and 1,200
mA shows severe artifact, precluding useful examination of ablation zone and
preventing procedural monitoring.
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C —CT fluoroscopy images of electrode placement in 71-year-old man with
hepatocellular carcinoma. Image obtained with 120 kVp, 40 mAs, and 8-mm slice
thickness during radiofrequency shows that increasing generator current from
1,200 to 2,000 mA has not appreciably changed artifact.
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1D —CT fluoroscopy images of electrode placement in 71-year-old man with
hepatocellular carcinoma. Image obtained with 120 kVp, 40 mAs, and 8-mm slice
thickness during radiofrequency pulsing (i.e., with no current flowing) shows
no artifact.
|
|
View larger version (105K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1E —CT fluoroscopy images of electrode placement in 71-year-old man with
hepatocellular carcinoma. Image obtained with 120 kVp, 250 mAs, and 8-mm slice
thickness during radiofrequency ablation with 200-W, 480-kHz generator, and
2,000 mA shows that increasing CT tube current from 40 to 250 mAs has lessened
artifact, compared with that in C.
|
|
Discussion
CT has been shown to be an effective method for monitoring radiofrequency
ablation. It can provide acceptable liver-to-tumor contrast and enables
accurate assessment of tumor size
[3]. The gas produced during
tissue coagulation is well visualized and confirms successful heating.
Furthermore, CT enables easy reference to the preablation scans, increasing
confidence that the tumor margins are being adequately ablated. CT has also
been shown to reflect ablation size more accurately (as determined by
pathologic examination) than does sonography
[4].
To meet the demands of interventional radiologists, all major vendors now
provide CT fluoroscopy equipment. CT fluoroscopy, a tool designed to shorten
procedures while still providing the excellent spatial localization of CT, was
originally introduced in 1993 by Toshiba Medical Systems on its Xpress/SX CT
scanner [5]. The meaning of
"CT fluoroscopy" varies, but the term is generally used to
describe the application of real-time CT in interventional procedures
[6]. Many radiologists,
including the authors, occasionally like to acquire CT fluoroscopic images
during the application of radiofrequency energy to confirm electrode placement
(i.e., to make sure that the electrode has not moved), examine for gas
formation, and look for the changes in tissue attenuation that have been
reported to occur with successful ablation of tissue
[3].
For the patient presented in this report, we acquired multiple sequential
images using 120 kVp, 40 mAs, an 8-mm slice thickness, a 0.5-sec tube rotation
time, and a 512 x 512 matrix, and we saw the artifact only during active
radiofrequency application. This finding was also observed in a companion case
of radiofrequency ablation using another 16-MDCT imaging platform (Aquilon 16,
Toshiba America Medical Systems).
The artifact encountered during these procedures resembled a beam-hardening
artifact but was unlikely to be such because it occurred only during active
radiofrequency ablation. Thus, we speculate that, similar to what occurs
during MRI, electromagnetic crosstalk occurred during active radiofrequency
pulsing and interfered with data acquisition in an as yet undetermined way.
Fast reconstruction times for real-time CT fluoroscopy often are achieved
using a number of simplifications in the reconstruction process, including
omitting image calculations such as beam-hardening corrections. Hence, it is
possible that the additional radiofrequency signal induces distortions, which
may be amplified by the simplified reconstruction algorithms, dramatically
reducing imaging quality. In addition to the previously mentioned
reconstruction algorithms, the thinner detectors of the multidetector
platforms may be more susceptible to interference.
Our observation raises many questions: We have observed this phenomenon on
only one manufacturer's CT units (although on two separate imaging platforms,
Aquilon 16 and Aquilon 64 [Toshiba America Medical Systems]) and using only
one type of radiofrequency electrode. It is unclear whether the artifact
reflects a specific compatibility problem between these two devices or
reflects a more widespread problem. In addition, and of greater concern, is
the possibility that the issue of reciprocal interference needs to be
addressed—that is, that using CT fluoroscopy during a radiofrequency
procedure might perturb the radiofrequency pulse generated. If this were the
case, the ability of a system to generate sufficient energy to heat the tumor
may be seriously limited.
References
- Dupuy DE, Goldberg SN. Image-guided radiofrequency tumor ablation:
challenges and opportunities—part II. J Vasc Interv
Radiol 2001; 12:1135
-1148[Medline]
- Goldberg SN, Stein MC, Gazelle GS, Sheiman RG, Kruskal JB, Clouse
ME. Percutaneous radiofrequency tissue ablation: optimization of
pulsed-radiofrequency technique to increase coagulation necrosis. J
Vasc Interv Radiol 1999; 10:907
-916[Medline]
- Cha CH, Lee FT Jr, Gurney JM, et al. CT versus sonography for
monitoring radiofrequency ablation in a porcine liver.
AJR 2000; 175:705
-711[Abstract/Free Full Text]
- Goldberg SN, Gazelle GS, Compton CC, Mueller PR, Tanabe KK.
Treatment of intrahepatic malignancy with radiofrequency ablation:
radiologic-pathologic correlation. Cancer2000; 88:2452
-2463[CrossRef][Medline]
- Katada K, Kato R, Anno H, et al. Guidance with real-time CT
fluoroscopy: early clinical experience. Radiology1996; 200:851
-856[Abstract/Free Full Text]
- Keat N. Real-time CT and CT fluoroscopy. Br J
Radiol 2001; 74:1088
-1090[Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
Y. Minami, H. Chung, M. Kudo, S. Kitai, S. Takahashi, T. Inoue, K. Ueshima, and H. Shiozaki
Radiofrequency Ablation of Hepatocellular Carcinoma: Value of Virtual CT Sonography with Magnetic Navigation
Am. J. Roentgenol.,
June 1, 2008;
190(6):
W335 - W341.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Introduction
Case Report
