DOI:10.2214/AJR.07.3698
AJR 2008; 191:790-792
© American Roentgen Ray Society
Summation of CT Scans During Radiofrequency Ablation for Assessing Target Lesion Coverage
Pamela M. Mazurek1,
Anne M. Silas2 and
John M. Gemery2
1 Department of Radiology, CT Scan Division, Dartmouth-Hitchcock Medical Center,
Lebanon, NH.
2 Department of Radiology, Division of Vascular and Interventional Radiology,
Dartmouth-Hitchcock Medical Center, One Medical Center Dr., Lebanon, NH
03756.
Received January 21, 2008;
accepted after revision March 27, 2008.
Address correspondence to J. M. Gemery
(john.m.gemery{at}hitchcock.org).
Abstract
OBJECTIVE. In radiofrequency ablation of lesions that require probe
repositioning, distinguishing between treated and untreated regions can be
difficult. We describe a method of using existing CT equipment to summate
images of a current probe placement with those of earlier placements or scans
of transiently enhancing targets.
CONCLUSION. Summation of CT scans during radiofrequency ablation
involving multiple probe placements is feasible and results in a better
appreciation of probe positioning relative to the target lesion.
Keywords: CT • CT scan • CT summation • probe placement • radiofrequency ablation
Introduction
Treatment of lesions with radiofrequency ablation may require probe
repositioning during the procedure because the lesion may be larger than the
probe or have a shape that does not correspond with available probe geometry.
Determining the extent of ablation coverage of the target lesion may thus be
challenging, particularly because the heating does not produce a visible
change in the treated region during the procedure. We did not discover in the
literature an existing method for merging (summating) into a single display CT
images of current probe placement with images of one or more probe placements
from earlier in the radiofrequency ablation procedure. In the absence of
summated images, the operator must, during the course of the radiofrequency
ablation procedure, serially review scans of different probe placements to
assess the adequacy of target lesion coverage. This may be more than an
inconvenience to the operator because studies have shown that the degree of
complete target necrosis diminishes with increasing target size
[1,
2].
We describe a rapid method for summating CT scans of different probe
placements into a single set of images, providing better appreciation of all
tine positions relative to the target lesion. Probe tines from different
placements all appear in the same set of images, as if all probe placements
were simultaneous. This method also permits targeting of lesions that enhance
only briefly. The summation method we describe is applicable to all GE
Healthcare CT equipment.
Technique
Scanner options, inputs, and selections described below are for the HiSpeed
CT/i system (GE Healthcare), a single-detector scanner with fluoroscopic
capability. The radiofrequency ab lation patient must be scanned using the
same parameters in each series to allow summation. At our institution,
patients are treated under general anesthesia, which is important for
obtaining accurate summation scans because respiration can be suspended during
the time in which each scan is acquired. Scanning data are acquired in the
helical mode only at 3-mm slice thickness (or 2.5 mm on GE Healthcare 16- and
64-MDCT scanners). Start and completion locations must remain constant across
the series, as must the display field of view and the tissue algorithm.
We typically perform initial probe placements with CT fluoroscopy or, if CT
fluoroscopic visu alization is insufficient, with axial scans reconstructed at
3-mm-thick slices at 3-mm intervals. Once the probe is considered to be in
final position for treatment, helical axial images are obtained using the
above parameters. Scans are reconstructed and displayed at 3-mm-thick slices
at 1-mm intervals for final recording of the probe placement. When the size or
shape of the target lesion requires probe repositioning, this process will be
repeated, except that such additional placements can be more precisely
targeted before treatment because of the image summation method described
here.
In preparation for merging different imaging series, the data set is
reconstructed to 3-mm slice thickness at 1-mm intervals. The merging process
is as follows. The scanner screen has a bottom line with a
"Retro-Recon" box, select it and a list of patients appears.
Choose the appropriate patient and to the right, a list of series appears.
Highlight the appropriate series and choose the "Select" box at
the bottom of the screen. A new screen appears that includes an "Image
Interval" box, and the number there should be set at 1.0. To the right
is a box containing the scanning parameters, which must match across the
series in order for all series to be combined. Choose "Accept" and
the com puter will reconstruct the scanning data to images with 3-mm thickness
at 1-mm intervals.
View larger version (142K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —Summated images from radiofrequency ablation procedures show
multiple individual probe placements as single images. Summated CT images show
liver lesion ablations in 55-year-old woman with metastatic leiomyosarcoma
(A), 70-year-old man with metastatic colon cancer (B), and
60-year-old man with hepatocellular carcinoma (C).
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —Summated images from radiofrequency ablation procedures show
multiple individual probe placements as single images. Summated CT images show
liver lesion ablations in 55-year-old woman with metastatic leiomyosarcoma
(A), 70-year-old man with metastatic colon cancer (B), and
60-year-old man with hepatocellular carcinoma (C).
|
|
View larger version (144K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C —Summated images from radiofrequency ablation procedures show
multiple individual probe placements as single images. Summated CT images show
liver lesion ablations in 55-year-old woman with metastatic leiomyosarcoma
(A), 70-year-old man with metastatic colon cancer (B), and
60-year-old man with hepatocellular carcinoma (C).
|
|
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1D —Summated images from radiofrequency ablation procedures show
multiple individual probe placements as single images. Summated CT image shows
renal ablation in 83-year-old man with renal cell carinoma.
|
|
View larger version (146K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1E —Summated images from radiofrequency ablation procedures show
multiple individual probe placements as single images. Summated image from
three series in 55-year-old man with hepatocellular carcinoma shows early
arterial phase showing transient enhancement of target lesion, initial probe
placement, and later probe placement.
|
|
To begin merging two series, select the box "Image Works" in
the initial screen on the scanner console. Choose the appropriate patient and
then select the first of the series to be combined. Highlight that series and
a list of all images in that series will appear and they should also be
highlighted. Then select the "Add/Sub" button, press "Select
Set" and the first series will appear in the form of an equation. There
is a space for one series, a plus symbol, a space for a second series, and an
equals symbol. One may then select the second series to be combined, and that
series will appear in the second box in the above equation. Before hitting the
equals button, be sure to set the series number in the box to the right to a
unique identifying number. When the equals button is touched, the new combined
series will save to the list of other series under that patient's examination
and with the unique identifying number that you have chosen. Once the final
series containing all probe placements has been created, that series can be
treated as any other set of images; it may be reformatted into sagittal or
coronal planes for better appreciation of tine positions relative to the
target lesion.
The reformation process will only merge two series, but because the
summated result of a merged series may be treated as any other set of images,
it is possible to combine it again with any number of additional probe
placement series, resulting in a summation in which three or more probe
placements may be seen. Select the third probe placement series, and then
select the new summation series of the previous two probe placements added
together, making sure that the current "Save State Series" is
different from the previous two. This process is repeated until all probe
placements are combined into one series. This final summation series may then
be reformatted into coronal and sagittal planes.
Discussion
Livraghi et al. [1,
2] have reported a lower rate
of complete target necrosis after radiofrequency ablation with increasing size
of the target lesion. This is likely due in part to the difficulty in
assessing total target lesion coverage when the operator must reposition the
probe. This repositioning always includes some degree of estimation because on
unenhanced CT the treated regions do not visibly differ from the un treated
regions. We have found the sets of images incorporating the target lesion and
multiple probe placements to be helpful in assessing target coverage. The
merged data may also be reformatted into planes other than axial, which may
improve probe placement (Fig.
1A,
1B,
1C,
1D,
1E). Antoch et al.
[3] compared axial and
multiplanar imaging of radio frequency ablation probe positioning during
hepatic ablations. They found that review of multiplanar images improved the
accuracy of probe placement versus reliance on axial images alone and have
advocated "multiplanar reformations to verify an optimal probe location
after axial CT-guided needle place ment" with radiofrequency ablation of
liver lesions.
Fujioka et al. [4] reported
merging before and after radiofrequency ablation CT images of patients treated
for hepatocellular carcinoma to assess target lesion coverage. Their method,
however, involved transfer of data to a workstation for the fusion process.
The method we describe does not require transfer to a workstation and makes
use of (in our institution) existing software and hardware. The merging
process is rapid enough that the merged images are used during the procedure
and allow intraprocedure corrections for better target coverage. For example,
merging two series of 46 images each requires approximately 30 seconds on a
single-detector scanner. The time needed for summation does increase with an
increasing number of slices to be merged but is also faster on newer scanner
models with greater processing power. Rapid merging of series may be
particularly useful with lesions that enhance only briefly. Subsequent probe
place ments may be compared with the lesion displayed in maximum
enhancement.
Finally, our use of general anesthesia and suspended respiration scanning
permits merging of data from any anatomic site. Although not all practitioners
perform ablations using general anesthesia, we think that for our practice the
benefits of suspended respiration and image summation outweigh the
disadvantages of general anesthesia.
The process described in this study is limited to the GE Healthcare family
of CT scanners. In 16- and 64-MDCT scanners, the computer will reconstruct the
scanning data to images of 2.5 mm rather than the 3-mm thickness described
here, but in other respects the process is the same throughout the GE
Healthcare line. Our results show the general feasibility of a radiofrequency
ablation summation technique that provides, in our view, better appreciation
of tine location relative to the target lesion and facilitates targeting of
lesions that are conspicuous only during a brief period of enhancement. Until
such time as devices are available in which a more automatic sum mation
process is offered, this method can help simplify assessment of lesion
coverage and improve accuracy of radio frequency ablations.
References
- Livraghi T, Goldberg SN, Lazzaroni S, Meloni F, Solbiati L, Gazelle
GS. Small hepatocellular carcinoma: treatment with radio-frequency ablation
versus ethanol injection. Radiology 1999;210
: 655–656[Abstract/Free Full Text]
- Livraghi T, Goldberg SN, Lazzaroni S, et al. Hepatocellular
carcinoma: radio-frequency ablation of medium and large lesions.
Radiology 2000;214
: 761–768[Abstract/Free Full Text]
- Antoch G, Kuehl H, Vogt FM, Debatin JF, Stattaus J. Value of CT
volume imaging for optimal placement of radiofrequency ablation probes in
liver lesions. J Vasc Interv Radiol 2002;13
:1155
–1161[CrossRef][Medline]
- Fujioka C, Horiguchi J, Ishifuro M, et al. A feasibility study:
evaluation of radiofrequency ablation therapy to hepatocellular carcinoma
using image registration of preoperative and postoperative CT. Acad
Radiol 2006; 13:986
–994[CrossRef][Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
