Key words
brachytherapy - interventional procedures - ultrasound computed tomography (US/CT)
Introduction
CT-guided interstitial brachytherapy (HDR-BT) is a catheter-based procedure that,
among other things, allows local ablation of thoracic and abdominal malignancies.
The irradiation catheter is usually positioned percutaneously using CT fluoroscopy.
Unlike thermal ablation procedures, such as radiofrequency ablation (RFA) and microwave
ablation (MWA), this treatment method has no technical limitation regarding tumor
size and proximity to heat-vulnerable structures [1]
[2]. Furthermore, interstitial irradiation is not subject to any thermal cooling effects
by adjacent vessels and has no influence on respiratory excursion due to the fixed
placement of the irradiation catheter in the tumor in relation to stereotaxis (SBRT)
[3]
[4].
Multiple catheters need to be introduced for multiple and larger lesions resulting
in a complex procedure typically associated with higher radiation exposure for the
radiologist [5]
[6]. Furthermore, the accessibility, especially of smaller lesions, under CT fluoroscopy
is often difficult and requires experience to reach the target lesion in a dose-saving
manner [7]
[8].
In thermal ablation procedures, such as radiofrequency ablation, in addition to computer
tomography, ultrasound in image guidance has become established and thus represents
an alternative to ionizing radiation [9].
In this study we report initial experiences with sonographically-assisted catheter
positioning in the interstitial brachytherapy of hepatic and renal malignancies as
a supplement to CT fluoroscopy.
Materials and Methods
Patient Cohort
Since July 2017 patients have been recruited and included in a prospective feasibility
study. The inclusion criteria include patients with planned, interstitial brachytherapy
for the ablation of tumors in sonographically clearly visible organs such as the liver
and kidneys. The study was reviewed and approved by the local ethics committee. Prior
to the procedure all patients provided their informed consent to study-specific activities
and approved the further processing of their clinical and radiographic data for study
purposes in accordance with data protection guidelines.
Twelve patients were included (8 men, 4 women, mean age 70 years) with a total of
16 tumors to be treated (colorectal liver metastases n = 9, renal cell carcinomas
n = 3, hepatocellular carcinomas n = 2 and cholangiocarcinomas n = 2).
Three patients had prior liver surgery, one patient with a trisectectomy and two patients
with atypical resection. Of the 13 liver lesions treated, five were local recurrences
after radiofrequency ablation or atypical resection.
To objectify patient selection, a pre-interventional evaluation of sonographic accessibility
was dispensed with.
Sonographically-assisted CT-supported Brachytherapy
The catheter placement was performed in an 80-row CT unit (Aquillion Prime, Canon
Medical Systems, Neuss, Germany) with concomitant analgosedation of the patient with
on-demand, intravenous administration of fentanyl and midazolam under pulse oximetry
monitoring. The interstitial access to the target lesion was performed via an initial,
image-guided puncture (if possible, sonographically-guided, otherwise performed CT-fluoroscopically)
with an 18Ga coaxial needle and the subsequent change to 25cm-long 6F catheter sheaths
(Terumo Radifocus® Introducer II, Terumo Europe, Leuven, Belgium) over a stiff guidewire (Amplatz SuperStiff
™, Boston Scientific, Marlborough, USA). A 6F irradiation catheter (afterloading catheter,
Primed® Medizintechnik GmbH, Halberstadt, Germany) was placed flush with the inner lumen
of the sheath and the system was fixed with a skin suture.
If the lesions were small and round (< 4 cm), an irradiation catheter was inserted
into a central position. In the case of a larger or irregularly shaped lesion, multiple
catheters were inserted to match the shape of the ablation zone (depending on the
access path in a fanned or crossed arrangement).
Once the patient was brought into the radiotherapy site, treatment planning was carried
out using a planning CT (Oncentra® Brachy, Elekta Instrument AB, Stockholm, Sweden) and single fraction irradiation
with an iridium 192 source used in afterloading technique. After defining the gross
tumor volume (GTV) based on the available image information, a safety margin of 5 mm
was added for the computer-assisted generation of the clinical target volume (CTV).
Due to the stable catheter position in the target volume, the CTV could then be equated
with the final planning target volume (PTV). Depending on the tumor entity, a target
dose of 15 Gy (renal cell carcinoma, hepatocellular carcinoma), 20 Gy (cholangiocarcinoma
carcinoma) or 25 Gy (colorectal liver metastasis) was prescribed for CTV/PTV.
Within the study, CT fluoroscopy (120 kVp / 30 mAs, 0.5 s rotation time, 6 mm single-slice
acquisition, image matrix 512 × 512) was replaced by laterally-positioned sonography
using low-frequency convex (1 – 5 MHz) and matrix (1 – 6 MHz) ultrasound heads (EPIQ7,
Philips Medical Systems, Amsterdam, The Netherlands) during the initial puncture and
interim position monitoring as often as technically feasable (see [Fig. 1]). The free-hand puncture technique was used for eight lesions, while four lesions
were punctured via a coaxial guide on the ultrasound head.
Fig. 1 Arrangement in the CT room for ultrasound-assisted punctures.
At the end of each procedure, contrast-enhanced computed tomography was performed
as needed to perform radiotherapy planning.
Two specialists in radiology with 7 and 4 years experience in percutaneous interventions
(at least 1000 and 300 documented percutaneous interventions, respectively) were responsible
for the performance and assessment of the interventions.
Study Design and Statistics
Patient characteristics, the number of catheters per imaging modality, intervention
and lesion, lesion parameters, and fluoroscopy times were tabulated. The image datasets
were recorded for each intervention performed, and the visibility of the lesions by
the intervention radiologists involved was assessed by consensus using a grading scale.
In addition, the dose information (CTV, target dose, D100) of all lesions treated
was collected from the treatment planning system.
The Society for Interventional Radiology (SIR) classification was used to evaluate
major and minor complications [10].
The collected data were first descriptively evaluated in SPSS 24.0 (IBM® SPSS® Statistics, IBM Deutschland GmbH, Ehningen, Germany) with determination of mean and
standard deviation as well as median and spread. Box plots were used to illustrate
the data. If a comparison of statistical variables between CT and ultrasound imaging
modalities was methodically feasible in the small patient population, this was done
by the Mann-Whitney U test for independent samples and the Wilcoxon signed-rank test.
Results
Image-guided Catheter Positioning
A total of 16 tumors with a mean diameter of 3.9 ± 2.7 cm (min 1.5 cm to max. 12.9 cm)
were treated using 2.3 ± 1.5 irradiation catheters (1 to 5 catheters per lesion, 28
catheters in total).
Catheter positioning could be completely achieved under ultrasound guidance in 12
of 16 lesions and 23 of 28 catheters. In 4 tumors, the initial puncture had to be
performed under CT fluoroscopy due to insufficient sonographic conditions. One liver
lesion was directly beneath the diaphragm at a resection margin after trisectorectomy
and another directly in the liver hilus. In 2 other tumors, sonographic visibility
was not considered sufficient for an accurate puncture.
On the whole, all punctures and catheter placements of kidney tumors and liver tumors
in the caudal segment row (segment 3/4B/5/6) were successfully performed sonographically.
In one patient with a lesion not visible in CT, sonographically-assisted puncture
and catheter placement completely replaced CT fluoroscopy (see [Fig. 2]). Here, only computed tomography with contrast agent application was necessary for
radiation planning.
Fig. 2 Interstitial brachytherapy of a subcapsular HCC in liver segment III. The lesion
is barely visible in non-enhanced CT A although being detected in previous MRI. Sonography B demonstrates a good visualization of the HCC as a hypoechogenic mass (arrows) and
is easily punctured (arrow head depicting the needle tip). Contrast-enhanced CT in
the arterial phase for irradiation planning C showing a central location of the catheter in the HCC (circle).
The planned target volume (CTV) dose was achieved in 14 of 16 lesions. In two cases
the dose was reduced due to the proximity of the tumor treated to risk organs (gall
bladder, maximum dose of 20 Gy, stomach / duodenum, maximum dose of 14 Gy).
Fluoroscopy Time and Lesion Visibility
In some cases of ultrasound-guided intervention, interim catheter placement controls
were performed using CT fluoroscopy (120 kVp / 30 mAs, 0.5 s rotation time, 6 mm single-slice
acquisition, image matrix 512 × 512). Mean fluoroscopy time for otherwise sonographically-guided
procedures, however, was significantly shorter (p = 0.006, see [Fig. 3]) at 14.5 s versus 105.5 s when CT fluoroscopy was used for the whole procedure.
Fig. 3 Mean fluoroscopy time (± standard deviation) for ultrasound-assisted puncture (Sonography
± CT) vs. CT puncture alone (CT) during catheter placement.
The visibility of the target lesions was assessed based on the consensus of the two
radiologists for both imaging modalities. In sonography, recognizability was rated
as very good in 8 out of 16 lesions based on graded assessments; in CT fluoroscopy,
this was only true for 2 lesions. Four or seven lesions were graded as good, two or
five were considered satisfactory. In 2 tumors there was deficient detectability in
the ultrasound (two cholangiocarcinomas) or CT (one hepatocellular carcinoma, one
colorectal liver metastasis). Other grades were not issued in the low number of cases.
[Fig. 4] provides an overview of the assessment of lesion visibility. Statistically, there
was no significant difference between the modalities (p = 0.27), although visibility
was better in sonography compared to CT in a total of 6 lesions.
Fig. 4 Visibility of the target lesion (n = 16) by imaging modality (sonography vs. CT).
Interstitial Tumor Ablation
The tumor-enclosing dose during single fraction irradiation was set at 15 to 25 Gy,
depending on the tumor entity, and the mean target dose was 20.6 ± 4.0 Gy. With respect
to the clinical target volume (CTV), the final dose distribution (D100) reached the
target dose in 14 out of 16 cases, averaging 19.3 ± 4.8 Gy. In two patients, the reason
for the reduced dose was the proximity of the tumor to neighboring radiation-sensitive
organs (stomach n = 1, duodenum n = 1).
Complications
After removal of the catheter sheaths, a sonographic or CT check was carried out after
approximately two hours to rule out acute hemorrhaging in all cases.
At the 30-day follow-up, no major or minor complications were observed in the patient
population after sonographically-assisted or direct CT fluoroscopic catheter placement.
Discussion
To the best of our knowledge, this feasibility study was the first to utilize sonography
during image-guided interstitial HDR brachytherapy of hepatic and renal tumors as
an image guidance modality for the initial puncture and catheter insertion. Previously
only CT or MRI fluoroscopy were used [11]
[12].
In an initial exploratory analysis, it was shown that a majority of catheter placements
for CT-guided HDR brachytherapy can be performed with sonography equipment additionally
positioned adjacent to the CT table. The kidneys as well as the caudal liver segments
(3/4B/5/6) appeared to be particularly suitable as sonographically-accessible regions;
in the previous patient cohort only a few lesions in one of cranial liver segments
(2/4A/7/8) were inaccessible. Here the results are in line with studies that, for
example, have assessed the value of ultrasound and CT for radiofrequency ablation
of hepatocellular carcinomas and which were able to document comparable results [9]. However, the benefits of CT fluoroscopy are also known when, similar to our cohort,
certain regions of the abdomen are difficult for ultrasound to access [13]
[14]. However, many percutaneous procedures in interventional radiology still lack a
comparative, randomized study between sonography and CT fluoroscopy.
The significance of the study is primarily limited by the small number of patients
on whom the possibilities of the novel technique was observed. The goal should now
be to use a larger number of cases to define the value of ultrasound-assisted catheter
placement for a practical implementation in CT-guided HDR brachytherapy. A suitable
comparison criterion appears to be the reduction of fluoroscopy time during CT, which
is proportional to the radiation exposure of the medical staff involved and could
already be significantly reduced in the present study with the aid of ultrasound [15]. Similarly, the immediate availability of a second imaging modality improves the
visibility and accessibility of certain lesions similar to the principle of CT / ultrasound
image fusion [16]. Ultimately, this may result in improved positioning of the irradiation catheters
or a reduction in the number needed for sufficient irradiation. For radiotherapy planning,
however, CT imaging will continue to be required, and the procedure will therefore
not be fully within the field of sonography.
Summary
Sonographically-assisted catheterization in interstitial HDR brachytherapy has the
potential to reduce the use of CT fluoroscopy and therefore the radiation exposure
of the interventional radiologist. The visibility of the target lesion in sonography
is in some cases superior to CT fluoroscopy and allows accurate catheter placement
even in previously operated patients.
The approaches gained from this study are intended to develop the concrete added value
of the procedure in subsequent investigations on a larger group of patients.
Central Statements / Clinical Relevance
-
Ultrasound-assisted catheterization during CT-guided brachytherapy of abdominal tumors
is technically feasible and safe.
-
Ultrasound-based puncture can improve catheter placement.
-
A significant reduction in fluoroscopy time can help reduce the radiation exposure
of medical personnel.
