Published online on 31 August 2012 Acta Radiol, doi: 10.1258/ar.2012.

120174

Original article

Dual-phase CT angiography through the port-catheter system

for hepatic arterial infusion chemotherapy using multislice
CT: assessment of system dysfunction and impact on
predicting clinical problems

Hiroshi Seki, Toshirou Ozaki and Hiroyuki Ooi

Department of Diagnostic Radiology, Niigata Cancer Center Hospital, Niigata, Japan
Correspondence to: Hiroshi Seki. Email: hseki@niigata-cc.jp

Abstract
Background: Hepatic arterial infusion (HAI) chemotherapy is being explored for treatment of malignant liver
tumors. Maintenance of HAI systems is important for effective treatment.
Purpose: To prospectively evaluate the efficacy of dual-phase CT angiography through the port-catheter
system for HAI chemotherapy.
Material and Methods: This study enrolled 47 patients receiving HAI chemotherapy for malignant liver
tumors who underwent dual-phase CT angiography through the port-catheter system using multislice CT.
Using maximum intensity projection images reconstructed from vascular-phase CT imaging, hepatic arterial
patency and catheter location were assessed. Using a combination of vascular- and perfusion-phase CT
imaging, system dysfunction and clinical problems were evaluated.
Results: Dual-phase CT angiography was conducted 156 times. Stenosis and obstruction of the hepatic
artery and catheter dislodgment were observed seven times in four patients and four times in three patients,
respectively. Diagnostic accuracy using vascular-phase CT imaging was 100%. In addition, development of
collateral blood supply to the liver and extrahepatic perfusion to the stomach were observed three times in
three patients and twice in two patients, respectively. Overall, system dysfunction occurred 16 times in
12 patients, and system correction and treatment modification were required 11 times in 10 patients. In
assessing system dysfunction and predicting clinical problems, the accuracy of dual-phase CT imaging
was 100%.
Conclusion: Dual-phase CT angiography through the port-catheter system is helpful for assessing catheter
system dysfunction and predicting clinical problems in HAI chemotherapy.

Keywords: CT angiography, liver, catheters, interventional

Submitted March 7, 2012; accepted for publication July 9, 2012

The liver is one of the most common sites for cancer metas-
tases that result in signicant morbidity and mortality (1).
The efcacy of systemic chemotherapy for metastatic colo-
rectal cancer has improved considerably in recent years
using combinations of infusional 5-uorouracil and leucov-
orin with oxaliplatin or irinotecan (2, 3), and the addition of
biological agents such as bevacizumab, a monoclonal anti-
body targeting vascular endothelial growth factor, and
cetuximab, a monoclonal antibody against epidermal
growth factor receptor (4, 5). Despite this progress in sys-
temic treatments, long-term survival of patients with
unresectable liver metastases remains short. Thus, alterna-
tive treatment strategies, such as combinations of hepatic
arterial infusion (HAI) chemotherapy with systemic cyto-
toxic chemotherapy (6 8), and HAI chemotherapy using
newer agents (9 11), are being explored for the treatment
of liver metastasis. Additionally, recent studies have
reported that HAI chemotherapy results in an effective
response in patients with advanced hepatocellular carci-
noma (12 14). Thus, HAI chemotherapy is a promising
treatment strategy for patients with different types of liver
malignancies.

Acta Radiologica 2012: 1 9. DOI: 10.1258/ar.2012.120174

Copyright 2012 by the The Foundation Acta Radiologica
2 H Seki et al.
.................................................................................................................................................
For effective HAI treatment, proper maintenance of port-
catheter systems is of great importance, as inadequate drug
distribution and catheter-related problems may reduce treat-
ment response and increase local toxicity (15 18). For eval-
uating perfusion patterns of infused drugs, hepatic arterial
perfusion scintigraphy and CT angiography through the
port-catheter system have proven valuable. To properly
evaluate the patency of the hepatic artery and the position
of the indwelling catheter, digital subtraction angiography
(DSA) through the port-catheter system or conventional
angiography via a transfemoral approach is required
(19 21). Therefore, both perfusion analysis and angio-
graphy are essential components of proper follow-up for
HAI systems.
Recently, multislice CT has been developed as a non-
invasive imaging modality for evaluation of vascular
anatomy (22 26). When the multislice CT technique is
applied, CT angiography through the port-catheter system
can potentially be used not only for assessing perfusion of
infused drugs but also for evaluating the condition of the
hepatic artery and indwelling catheter. Based on this
concept, using a 64-channel multislice CT, we have designed
a dual-phase CT method that involves intra-arterial admin-
istration of contrast medium through the port-catheter
system. In this technique, vascular images via the port-
catheter system are reconstructed from thin-slice CT data
in the early enhancement phase, and conventional perfusion
images are acquired in the late enhancement phase. The
objectives of the present study were to prospectively evalu-
ate the efcacy of dual-phase CT angiography through the
port-catheter system and to assess its clinical impact on pre-
dicting system correction and treatment modication in
patients receiving HAI chemotherapy.
Material and Methods
Eligibility criteria
All patients were required to have unresectable malignant
liver tumors, implanted port-catheter systems for HAI che-
motherapy, and a performance status of  2 according
to the Eastern Cooperative Oncology Group classication
(27). Hematologic and chemistry parameters were required
to be within the following limits: white blood cell
count  3000/mm3, serum total bilirubin , 3.0 mg/dL,
and serum creatinine  1.5 mg/dL. Minor extrahepatic
disease conrmed by radiological examination or intra-
operative ndings was not considered an absolute contra-
indication if the liver was the predominant disease site.
All patients provided informed consent prior to com-
mencement of the examination. The study was approved
by the ethics committee at our institution.
Catheter implantation procedure and HAI chemotherapy
Before implantation of the port-catheter system, angio-
graphy was performed for arterial road-mapping and arter-
ial redistribution. When an aberrant hepatic artery was
encountered, hepatic arterial blood ow was redistributed
using coils (FPC35 Pt-Max and VortX; Boston Scientic,
Natick, MA, USA; Tornado; Cook, Bloomington, IN, USA)
to convert multiple hepatic arteries into a single arterial
blood supply. If necessary, extrahepatic arterial branches
arising from the hepatic artery and parasitic arteries with
hepatopetal blood ow were also occluded using coils
and/or a mixture of n-butyl cyanoacrylate (Histoacryl;
Braun, Melsungen, Germany) and iodized oil (Lipiodol;
Laboratorie Guerbet, Roissy, France). A port-catheter sys-
tem was then implanted during a surgical procedure via
the left axillary artery, or with a percutaneous approach
via the right femoral artery. A long, tapered side-hole-type
catheter composed of a 2.7-F, 20-cm-long distal shaft and
a 5-F, 70-cm-long proximal shaft (Anthron P-U catheter;
Toray, Tokyo, Japan: W-spiral catheter; Piolax Medical
Devices, Yokohama, Japan) was inserted in the hepatic
artery using the following methods: the distal shaft of the
catheter with a side-hole was placed in the common
hepatic artery and the tip of the catheter was xed in the
gastroduodenal artery using coils; the distal shaft was
inserted distally into the hepatic artery and its side-hole
was positioned proximally (28, 29). The proximal end
of the catheter was cut and connected to the port (Soph-
A-Port; Sophysa, Orsay, France), which was implanted in
the subcutaneous space. Chemotherapeutic agents were
infused into the hepatic artery through the side-hole of the
catheter.
HAI chemotherapy was initiated 2 to 14 days after cath-
eter placement. The following chemotherapeutic agents
were used: 5-uorouracil (1000 mg/m2 weekly as a 5-h infu-
sion) was administered to patients with liver metastases
from colorectal and testicular cancer; 5-uorouracil
(330 mg/m2 weekly as a 2-h infusion), mitomycin C
(2.7 mg/m2 biweekly as a bolus infusion), and epirubicin
(30 mg/m2 every 4 weeks as a bolus infusion) were adminis-
tered to patients with liver metastases from gastric, breast,
and gallbladder cancer; and 5-uorouracil (1000 mg/m2
weekly as a 5-h infusion) and cisplatin (6.7 mg/m2 weekly
as a bolus infusion) were administered to patients with
hepatocellular carcinoma and liver metastases from esopha-
geal cancer. Patients received HAI chemotherapy until one
of the following occurred: progressive liver disease, extra-
hepatic progression that was predominant over the hepatic
lesions, catheter dysfunction and/or complications that pre-
vented further treatment, or excessive toxicity. In the event
of a complete response, treatment was continued for an
additional 3 months.
Dual-phase CT angiography and DSA through the
port-catheter system
Patients were followed until the end of HAI chemotherapy.
Dual-phase CT angiography and DSA through the port-
catheter system were performed within 10 days after
implantation of the catheter-port system and every 2 to
3 months thereafter to assess patency of the hepatic artery,
position of the catheter, and perfusion pattern in the liver.
Imaging was also performed when patients complained
of any symptoms related to HAI of the chemotherapeutic
agents.

.................................................................................................................................................
Dual-phase CT angiography was performed using a
64-channel multislice CT scanner (Light Speed VCT; GE
Healthcare, Milwaukee, WI, USA). The following CT par-
ameters were used: a pitch of 0.984, a gantry rotation
speed of 0.5 s, a detector collimation of 64  0.625 mm,
120 kV, auto mA (max 650 mA), a matrix of 512  512,
and a eld of view of 35 cm. A 24-gauge Huber-type
needle (Coreless needle; Nipro, Osaka, Japan) was used
for access to the implantable port. The contrast medium
was diluted with normal saline to contain 100 mg I/mL of
iopamidol (Iopamiron; Bayer Schering Pharma, Osaka,
Japan), and 25 mL of the solution was injected via the
implantable port at a rate of 0.7 mL/s during the following
dual-phase CT scanning procedure. CT data acquisition
began 6 s after the initiation of injection, and multiple sec-
tions encompassing the entire liver were obtained in a cra-
niocaudal direction with an approximately 4-s scan time
during a single breath-hold to evaluate conguration of
the hepatic artery and vascular ow through the port-
catheter system. After allowing breathing for 16 s, CT
images were taken in the same range during an additional
breath-hold to assess perfusion of the contrast media in
the liver. The former and latter CT imaging was dened
as the vascular phase and the perfusion phase, respectively.
Contiguous axial images were reconstructed with a 5-mm
thickness. Additionally, for three-dimensional (3-D) post-
processing, 0.625-mm axial images were obtained from CT
data in each phase. In the vascular phase, maximum inten-
sity projection (MIP) reconstructions were created with a
1808 view parallel to the longitudinal axis of the patient. If
necessary, a volume-rendering technique was additionally
used. In the perfusion phase, reformatted CT images were
created in coronal and sagittal directions. The 3-D imaging
reconstructions were generated on a commercially available
workstation (Advantage Windows 4.3 and AW Sweet 2.0,
GE Healthcare).
DSA was performed using manual injection of the con-
trast medium through the port-catheter system to conrm
hepatic arterial patency and the position of the indwelling
catheter. Images were acquired within 10 days after dual-
phase CT angiography.
Image analysis of dual-phase CT angiography
Dual-phase CT angiography was assessed via a consensus
of two experienced radiologists who were unaware of clini-
cal data except for data about the catheter placement
method used for each patient.
Initially, in vascular-phase CT images, patency of the
hepatic artery and location of the indwelling catheter
were evaluated using reconstructed MIP imaging.
Volume-rendered imaging, original axial thin-slice CT
images, and reformatted coronal and sagittal CT images
were additionally used to assist evaluation. Patency of the
hepatic artery, ranging between the common hepatic
artery and proximal portions of the segmental hepatic
artery, was classied into three categories: obstruction,
stenosis, and normal patency. Stenosis was dened as
narrowing of the vascular lumen exceeding 50%. Catheter
dislodgment was assessed based on ow patterns and
Dual-phase CT angiography through the port-catheter system 3
position of the catheter tip in vascular-phase MIP
imaging, which was compared with an angiogram via the
port-catheter system obtained just after implantation.
Dislodgment was diagnosed as follows: when the splenic
artery was seen with normal patency of the hepatic artery
and the catheter tip was withdrawn, we judged that the
side-hole of the catheter was withdrawn from the hepatic
artery into the celiac arterial trunk; when the abdominal
aorta was seen and the catheter tip was withdrawn, we
judged that the side-hole of the catheter was withdrawn
into the aorta; when only either the right or left hepatic
artery was observed with no visualization of the proper
hepatic artery and the catheter tip was advanced, we
judged that the catheter was advanced distally into each
lobar hepatic artery. When the catheter tip was considered
to be appropriately located while the splenic artery was
seen with normal patency of the hepatic artery, we judged
that the catheter was not dislodged and that the contrast
medium had owed over the common hepatic artery into
the splenic artery due to an excessive injection rate or the
inuence of the cardiac cycle. When the ow of the contrast
medium via the port-catheter system was satisfactorily
observed while minor movement of the catheter tip was sus-
pected, it was managed as having no catheter dislodgment.
These interpretations were correlated with the ndings of
DSA through the port-catheter system, and the diagnostic
value of stenosis and obstruction of the hepatic artery and
catheter dislodgment was assessed.
Subsequently, in the combination vascular- and perfusion-
phase CT images, dysfunctions of the port-catheter system
were evaluated. Perfusion abnormalities via the port-catheter
system were assessed using axial, coronal, and sagittal
reformatted perfusion-phase CT images. If intrahepatic per-
fusion defects and/or extrahepatic perfusion were observed,
vascular-phase CT imaging was used to conrm hepatic
arterial obstruction, catheter dislodgment, and development
of extrahepatic vessels. When intrahepatic perfusion defects
were seen on perfusion-phase CT imaging with normal
patency of the hepatic artery on vascular-phase MIP
imaging, development of collateral blood supply to the liver
was suspected. When extrahepatic perfusion reaching the
stomach beyond the duodenum was observed on perfusion-
phase CT imaging with extension of many vessels into the
stomach and duodenum on vascular-phase MIP imaging,
system dysfunction potentially caused by gastroduodenal
ulcer or gastroduodenitis was presumed. When extrahepatic
perfusion was seen only in part of the duodenum on
perfusion-phase CT imaging, it was not considered to be
system dysfunction, as such perfusion was not necessarily
associated with gastroduodenal toxicity (30). These interpret-
ations were correlated with clinical symptoms and imaging
ndings of DSA through the port-catheter system, additional
angiography via the transfemoral route, and endoscopy,
all of which were performed to investigate catheter system
dysfunction and to resolve any problems.
Statistical analysis
In vascular-phase CT imaging, detection of hepatic arterial
stenosis and obstruction, as well as catheter dislodgment,
4 H Seki et al.
.................................................................................................................................................
was evaluated. With the DSA results used as the standard of
reference, the accuracy, sensitivity, specicity, positive pre-
dictive value, and negative predictive value were calculated.
The diagnostic value in assessing dysfunction of the port-
catheter system was analyzed using combined vascular- and
perfusion-phase CT imaging results. Based on the ndings
of DSA through the port-catheter system, conventional
angiography, and endoscopy, as well as clinical symptoms,
the accuracy, sensitivity, specicity, positive predictive
value, and negative predictive value were calculated.
Additionally, the impact of dual-phase CT angiography on
predicting clinical problems was evaluated. Clinical pro-
blems were dened as cases that actually required system
correction and treatment modication. Using cases with
actual clinical problems as the reference, the accuracy in pre-
dicting clinical problems was calculated.
Results
Between January 2007 and April 2009, 47 patients (age
range, 46 81 years; mean age, 63.6 years) at our institution
who received HAI chemotherapy using the port-catheter
system for malignant liver tumors were entered into this
study. Patient characteristics are listed in Table 1. Forty-
three patients had liver metastases from the following
primary tumors: colorectal cancer (n 20), gastric cancer
(n 14), esophageal cancer (n 4), breast cancer (n 3),
gallbladder cancer (n 1), and testicular cancer (n 1).
Four patients had hepatocellular carcinoma.
Table 1 Patient characteristics
Characteristic Patients (n)
Age (years)
,60 18
60 29
Gender
Men 34
Women 13
ECOG performance status
0 24
1 21
2 2
Liver tumors
Liver metastases 43
Hepatocellular carcinoma 4 using coils to convert the hepatic arterial blood ow into
Access route for catheter insertion
Left axillary artery 33
Right femoral artery 14
Side-hole catheter placement
in the hepatic artery
Method with fixing a catheter-tip 36
in the gastroduodenal artery
Method with inserting a distal catheter-shaft 11
distally in the hepatic artery

The distal shaft of a long tapered catheter with a side-hole was placed in the
common hepatic artery and the tip of the catheter was fixed in the
gastroduodenal artery using coils
The distal shaft of a long tapered catheter was inserted distally into the
hepatic artery and its side-hole was positioned proximally
ECOG, Eastern Cooperative Oncology Group; n, number of patients
Dual-phase CT angiography through the port-catheter
system was successfully performed during HAI chemother-
apy in all 47 patients. Overall, dual-phase CT angiography
was conducted 156 times (range, 1 10 times; mean, 3.3
times), with a mean follow-up period of 236 days (range,
57 741 days).
Evaluation of hepatic arterial patency and catheter location
using vascular-phase CT imaging
In all 156 CT examinations, hepatic arterial patency was cor-
rectly assessed using MIP imaging in the vascular phase;
these results were equivalent to those obtained using
DSA through the port-catheter system. Obstruction of the
proper hepatic artery was identied 5 months after
initiation of chemotherapy in one patient, and HAI che-
motherapy was subsequently terminated. Stenosis of the
hepatic artery was observed six times in three patients as
follows: stenosis of the common hepatic artery was seen
8 months after initiation of chemotherapy in one patient,
and HAI chemotherapy was discontinued after dissection
of the common hepatic artery was observed via reformatted
sagittal and coronal CT images (Fig. 1); stenosis of the
proper hepatic artery was seen 3 months after initiation of
treatment in one patient; and stenosis of both the right
and left hepatic arteries was observed at the initial examin-
ation and at 3, 6, and 9 months in subsequent exams in
one patient, which did not lead to discontinuation of HAI
chemotherapy. In the remaining 149 examinations, normal
patency was observed in the hepatic artery.
Catheter dislodgment was correctly evaluated using
vascular-phase MIP imaging. On four examinations in three
patients, catheter dislodgment could be detected using
vascular-phase CT imaging as well as DSA through the port-
catheter system. In one patient, withdrawal of the side-hole of
the catheter from the common hepatic artery to the celiac
arterial trunk was observed 9 months after initiation of che-
motherapy (Fig. 2), and the catheter was thus replaced. In
one patient, withdrawal of the catheter into the celiac arterial
trunk was seen 3 and 6 months after initiation of chemother-
apy, which required chemotherapy dosage reduction. In
another patient, advancement of the side-hole of the catheter
from the proper hepatic artery into the right hepatic
artery was observed 3 months after initiation of chemother-
apy; the left hepatic artery was subsequently embolized
the right hepatic artery. On seven examinations in four
patients, while the catheter tip was not withdrawn, the
splenic artery was observed with normal patency of the
hepatic artery on vascular-phase MIP imaging 3, 6, 9, and
10 months after initiation of chemotherapy in each patient,
and thereafter in two of these patients. We judged that the
catheter was not dislocated and that these were caused by
an overow of the contrast medium. Subsequently, using
DSA through the port-catheter system, it was conrmed
that these cases, in which the ow into the splenic artery
was not seen on the appropriate control by manual injection,
had no catheter dislodgment. On the remaining 145 examin-
ations, appropriate catheter location was conrmed using
vascular-phase CT imaging.
 Dual-phase CT angiography through the port-catheter system 5
.................................................................................................................................................

Fig. 1 An 81-year-old woman with stenosis and dissection of the common hepatic artery. (a) MIP and (b) volume-rendered images in the vascular phase show
stenosis of the common hepatic artery (arrow); (c) Stenosis of the common hepatic artery (arrow) is seen on DSA through the port-catheter system as well as in the
MIP and volume-rendered images; (d) A reformatted sagittal vascular-phase CT image shows dissection of the common hepatic artery (arrow)

Volume-rendered imaging in the vascular phase was
helpful to assess vascular conguration (Figs. 1 and 2). In
several cases, however, an imaging artifact was conspicu-
ously revealed around metallic coils in volume-rendering
reconstructions (Fig. 2), while less pronounced on MIP pro-
cessing; this may be a potential disadvantage for evaluating
patency of the hepatic artery around coils.
In assessing hepatic arterial patency and catheter location,
the accuracy, sensitivity, specicity, positive predictive
value, and negative predictive value of vascular-phase CT
imaging compared to DSA were all 100%. The diagnostic
value of vascular-phase MIP imaging was equivalent to
that of DSA through the port-catheter system.
Evaluation of system dysfunction using dual-phase
CT imaging
Assessment of catheter system dysfunction using a combi-
nation of vascular- and perfusion-phase CT imaging is
shown in Table 2. Dysfunction of the port-catheter system,
which was accurately identied using dual-phase CT angio-
graphy, occurred 16 times in 12 patients.
Using dual-phase CT angiography, obstruction and steno-
sis of the hepatic artery was accurately identied seven
times in four patients, in one of whom dissection of the
common hepatic artery was observed with stenosis of
the common hepatic artery (Fig. 1). Catheter dislodgment
was also accurately identied four times in three patients
(Fig. 2).
In three patients, development of collateral blood supply
to the liver, resulting in an intrahepatic perfusion defect in
the right subphrenic area, was suspected using perfusion-
phase CT imaging 2, 5, and 10 months after initiation of
chemotherapy. Subsequently, angiography via the transfe-
moral approach was performed, and the right inferior
phrenic artery was conrmed as a collateral vessel to the
liver (Fig. 3). It was then embolized using a mixture of
n-butyl cyanoacrylate and iodized oil to correct drug
distribution.
In two patients, a wide range of extrahepatic perfusion
was observed not only in the duodenum but also in the
antrum of the stomach on perfusion-phase CT imaging
6 and 12 months after initiation of chemotherapy.
Development of many gastroduodenal arterial branches
was also observed on vascular-phase MIP imaging
(Fig. 4). We judged these to be system dysfunction poten-
tially associated with gastric toxicity. Based on this
nding, gastric endoscopy was performed and gastric
ulcers and acute gastritis were diagnosed. One of these
patients complained of epigastric discomfort during HAI,
and angiography was subsequently performed showing
the development of collaterals to the stomach, which
were then embolized using coils. In another patient
who complained of anorexia, dosage reduction of che-
motherapeutic drugs was required. In the other 14 patients,
on 46 examinations, extrahepatic perfusion was observed
in part of the duodenum and the pancreatic head.
These were judged as cases without system dysfunction,
6 H Seki et al.
.................................................................................................................................................

Fig. 2 A 63-year-old woman with catheter dislodgment. (a) Angiogram via the port-catheter system obtained just after implantation shows a catheter tip inserted
distally into the gastroepiploic artery (black arrow), a side-hole of the catheter placed in the common hepatic artery (white arrow), and a distal catheter shaft fixed
within the gastroduodenal artery using coils (arrowheads); (b) In a vascular-phase MIP image obtained 9 months after initiation of chemotherapy, the splenic
artery (arrows) and the left inferior phrenic artery are seen with normal patency of the hepatic artery, and the catheter tip is withdrawn into the proximal
portion of the gastroepiploic artery (arrowhead); (c) In a vascular-phase volume-rendered image, the imaging artifact is conspicuously observed around metallic
coils (arrows) while vascular configurations are well demonstrated; (d) Subsequent DSA through the port-catheter system shows dislocation of the side-hole of the
catheter into the celiac arterial trunk; these findings are similar to those obtained using vascular-phase MIP imaging

and unresolved gastric toxicity was not observed during Discussion

treatment.
In assessing system dysfunction, the accuracy, sensitivity,
specicity, positive predictive value, and negative predic-
tive value of dual-phase CT angiography were all 100%.
Prediction of clinical problems using dual-phase
CT angiography
Clinical problems related to dysfunction of the port-catheter
system are presented in Table 3. In 10 patients with system
dysfunction, actual management was required for system
correction and treatment modication. Clinical problems
were correctly identied 11 times in these patients using
dual-phase CT angiography, and adjustment of the catheter
system and treatment modication was performed. In the
other 145 exams in which the port-catheter system was
deemed to be functioning correctly, no adjustments were
required during HAI chemotherapy. Thus, clinical problems
were accurately identied and assessed using dual-phase
CT angiography in all cases; the accuracy in predicting clini-
cal problems was 100%.
The present results demonstrate that dual-phase CT angio-
graphy through the port-catheter system is useful for
follow-up after administration of HAI chemotherapy via
the port-catheter system. This type of angiography can be
used not only to assess drug distribution, but also to evalu-
ate hepatic arterial patency and catheter location instead of
DSA through the port-catheter system.
In patients undergoing HAI chemotherapy for malignant
liver tumors, dysfunction of the port-catheter system
occasionally occurs, resulting in treatment failure.
Therefore, it is important to promptly identify and resolve
these problems. To evaluate sites of perfusion abnormalities
during HAI chemotherapy, CT angiography through the
port-catheter system is appropriate (20, 21). This is achieved
with intra-arterial administration of contrast medium
through the indwelling catheter and image acquisition at a
single phase in which the hepatic parenchyma is sufciently
enhanced to evaluate intrahepatic perfusion. However, this
conventional CT angiography is unsuitable for demonstrat-
ing the conguration of the hepatic artery due to poor con-
trast between vessels and enhanced hepatic parenchyma. In
 Dual-phase CT angiography through the port-catheter system 7
.................................................................................................................................................

Table 2 Assessment of system dysfunction using dual-phase CT enhanced hepatic parenchyma. In the present study,
angiography obstruction and stenosis of the hepatic artery were accu-
Identifications of rately evaluated using vascular-phase MIP imaging with
system dysfunction results equivalent to those obtained using DSA through
System using dual-phase
the port-catheter system. These results indicate that
dysfunction Patients (n) Events (n) CT (n)
vascular-phase CT imaging can be utilized to assess
Obstruction and 4 7 7 hepatic arterial patency instead of DSA through the port-
stenosis of the
hepatic artery
catheter system. Moreover, in the present study, catheter
Catheter 3 4 4 dislodgment was correctly identied on vascular-phase
dislodgment MIP imaging using ndings with an inow to the splenic
Development of 3 3 3 artery or aorta and dislocation of the catheter tip.
collateral blood
Although the contrast medium owed over the common
supply to the
liver hepatic artery into the splenic artery in several cases with
Development of 2 2 2 no catheter dislocation, this could be distinguished from
extrahepatic catheter dislodgment based on the position of the catheter
perfusion to the tip. As a result, the location of the side-holed catheter was
stomach
accurately assessed in all cases using vascular-phase MIP

In one of seven examinations, dissection of the common hepatic artery was imaging. We consider vascular-phase CT imaging to be a
also observed reliable method for assessing catheter dislodgment, similar
to DSA through the port-catheter system.
general, DSA through the port-catheter system has to be Volumetric 3-D rendering is a useful technique for
performed to assess hepatic arterial patency and catheter demonstrating vascular conguration. In the present
location. study, however, an imaging artifact was remarkably
Sone et al. rst reported that obstruction and stenosis of observed around metallic coils in volume-rendered
the hepatic artery were observed using MIP processing imaging in several cases, while it was less so on MIP
obtained from multislice CT angiography through the port- imaging. This artifact may become a potential drawback
catheter system (22). However, with this method, a 5-min in evaluating patency of the hepatic artery around coils, as
waiting period is required after the rst scan of conventional metallic coils are frequently used for catheter placement.
CT angiography to eliminate contrast materials in the Therefore, we consider that MIP processing is suitable for
hepatic parenchyma. In addition, visualization of the assessing patency of the hepatic artery in patients with port-
hepatic artery may be decreased due to the residual catheter systems.
enhancement in the hepatic parenchyma. In perfusion-phase CT imaging, which is the same pro-
In the dual-phase CT angiography through the port- cedure that is performed for conventional single-phase CT
catheter system used in the present study, vascular-phase angiography, sites of intra- and extrahepatic perfusion
CT imaging is performed 6 s after initiation of injection, fol- abnormalities can be clearly identied. In the present
lowed by perfusion-phase CT imaging at a 16-s delay after study, perfusion defects in the liver due to development
scanning the vascular phase. Thus, a long waiting period of collateral blood supply and extrahepatic perfusion in
is not required. On vascular-phase CT imaging, the the stomach resulting in gastric toxicity could be identied
hepatic artery was clearly revealed using MIP processing, using perfusion-phase CT imaging. Using a combination
with good contrast between the hepatic artery and the less of vascular- and perfusion-phase CT imaging, an

Fig. 3 A 61-year-old man with development of collateral blood supply to the liver. (a) A reformatted coronal perfusion-phase CT image shows an intrahepatic
perfusion defect in the right subphrenic area (arrows); (b) Angiogram of the right inferior phrenic artery shows collateral blood supply to the liver tumor (arrows)
8 H Seki et al.
.................................................................................................................................................

Fig. 4 A 63-year-old man with extrahepatic perfusion in the stomach resulting in gastric ulcer. (a) A perfusion-phase CT image shows extrahepatic perfusion in
the gastric antrum (arrows) beyond the duodenum (arrowhead); (b) A vascular-phase MIP image shows development of several gastroduodenal vessels (arrows)

Table 3 Prediction of clinical problems using dual-phase CT angiography

Identifications of
Clinical problems related clinical problems
to system dysfunction Patients (n) Management Events (n) using dual-phase CT (n)
Obstruction of the 1 Termination of HAI chemotherapy 1 1
proper hepatic artery and initiation of systemic therapy
Dissection of the common 1 Termination of HAI chemotherapy 1 1
hepatic artery with arterial stenosis and initiation of systemic therapy
Catheter dislodgment 3 Catheter replacement (n 1); 4 4
dosage reduction (n 1);
hepatic arterial
redistribution (n 1)
Intrahepatic perfusion defect due 3 Embolization of collaterals 3 3
to development of collateral
blood supply to the liver
Gastric ulcer and acute gastritis due 2 Embolization of gastric 2 2
to development of collaterals vessels (n 1); dosage
to the stomach reduction (n 1)

Clinical problems are defined as cases that actually required system correction and/or treatment modification

accurate rate (100%) of assessment of port-catheter system
dysfunction was obtained. Based on these ndings, clinical
problems were resolved using appropriate management via
the HAI system and treatment. In predicting clinical pro-
blems requiring system correction and treatment modi-
cation, the accuracy was 100% using dual-phase CT
angiography. These results indicate that dual-phase CT
angiography is greatly advantageous for administering
safe and effective HAI chemotherapy.
Several recent studies have reported that 3-D reconstruc-
tion from multislice CT data with intravenous adminis-
tration of contrast medium is useful for preoperative
evaluation of hepatic arterial anatomy (23 26). Using this
technique, hepatic arterial patency can potentially be
assessed. However, in intravenous enhanced CT angiogra-
phy, many arterial branches of abdominal organs overlap
the hepatic artery, which may hide abnormalities of the
hepatic artery. In addition, since the side-hole of the
implanted catheter is not visible and blood ow through
the port-catheter system is not depicted using this tech-
nique, the position of the side-hole of the catheter cannot
be evaluated. Therefore, we consider vascular-phase CT
imaging through the port-catheter system to be superior to
intravenous CT angiography in assessing the condition of
the hepatic artery and the indwelling catheter.
The present study has some limitations. First, in vascular-
phase MIP imaging, overow of the contrast medium may
cause false-positive judgment of catheter dislodgment in
cases with minor movement of the catheter that does not
inuence the drug distribution, although no such cases
were observed in the present study. Therefore, when the
splenic artery is observed with normal patency of the
hepatic artery on vascular-phase MIP imaging, the position
of the catheter tip should be carefully assessed. Second, an
intrahepatic perfusion pattern on perfusion-phase CT angio-
graphy may differ from the actual drug distribution, since
the injection rate using CT angiography is higher than that
used during continuous drug infusion. In the present
study, abnormal perfusion patterns on perfusion-phase CT
imaging were well correlated with actual clinical problems.
However, when a problem that cannot be explained by
CT angiography ndings is observed, addition of hepatic

.................................................................................................................................................
arterial perfusion scintigraphy should be considered. Third,
in the present study, the occurrence of dysfunction of the
port-catheter system was small. A larger number of these
events would be necessary to verify the present results.
In conclusion, this study demonstrated that dual-phase
CT angiography through the port-catheter system is a
reliable method for assessing dysfunction of the port-
catheter system and predicting system correction and treat-
ment modication in patients receiving HAI chemotherapy.
With regard to identication of stenosis and obstruction of
the hepatic artery and catheter dislodgment, vascular-phase
CT angiography can be utilized instead of DSA through the
port-catheter system. We believe that dual-phase CT angio-
graphy is a helpful follow-up examination in the manage-
ment of port-catheter systems.
Conict of interest: None.
REFERENCES
1 Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence
after hepatic resection for metastatic colorectal cancer: analysis of 1001
consecutive cases. Ann Surg 1999;230:309 18
2 Tournigand C, Andre T, Achille E, et al. FOLFIRI followed by FOLFOX6
or the reverse sequence in advanced colorectal cancer: a randomized
GERCOR study. J Clin Oncol 2004;22:229 37
3 Colucci G, Gebbia V, Paoletti G, et al. Phase III randomized trial of
FOLFIRI versus FOLFOX4 in the treatment of advanced colorectal
cancer: a multicenter study of the Gruppo Oncologico DellItalia
Meridionale. J Clin Oncol 2005;23:4866 75
4 Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and
cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal
cancer. N Engl J Med 2004;351:337 45
5 Giantonio BJ, Catalano PJ, Meropol NJ, et al. Bevacizumab in
combination with oxaliplatin, uorouracil, and leucovorin (FOLFOX4)
for previously treated metastatic colorectal cancer: results from the
Eastern Cooperative Oncology Group Study E3200. J Clin Oncol
2007;25:1539 44
6 Kemeny N, Jarnagin W, Paty P, et al. Phase I trial of systemic oxaliplatin
combination chemotherapy with hepatic arterial infusion in patients
with unresectable liver metastases from colorectal cancer. J Clin Oncol
2005;23:4888 96
7 Gallagher DJ, Capanu M, Raggio G, et al. Hepatic arterial infusion plus
systemic irinotecan in patients with unresectable hepatic metastases from
colorectal cancer previously treated with systemic oxaliplatin:
a retrospective analysis. Ann Oncol 2007;18:1995 9
8 Kemeny NE, Melendez FD, Capanu M, et al. Conversion to resectability
using hepatic artery infusion plus systemic chemotherapy for the
treatment of unresectable liver metastases from colorectal carcinoma.
J Clin Oncol 2009;27:3465 71
9 Ducreux M, Ychou M, Laplanche A, et al. Hepatic arterial oxaliplatin
infusion plus intravenous chemotherapy in colorectal cancer with
inoperable hepatic metastases: a trial of the Gastrointestinal Group of the
Federation Nationale des Centres de Lutte Contre le Cancer. J Clin Oncol
2005;23:4881 7
10 Boige V, Malka D, Elias D, et al. Hepatic arterial infusion of oxaliplatin
and intravenous LV5FU2 in unresectable liver metastases from colorectal
cancer after systemic chemotherapy failure. Ann Surg Oncol
2008;15:219 26
11 Tsimberidou AM, Letourneau K, Fu S, et al. Phase I clinical trial of
hepatic arterial infusion of paclitaxel in patients with advanced cancer
and dominant liver involvement. Cancer Chemother Pharmacol
2011;68:247 53
12 Kuroda M, Kobayashi Y, Urawa N, et al. Hepatic arterial infusion of
5-uorouracil in combination with subcutaneous interferon-alpha for
Dual-phase CT angiography through the port-catheter system 9
advanced hepatocellular carcinoma. Hepatogastroenterology
2007;54:518 21
13 Park JY, Ahn SH, Yoon YJ, et al. Repetitive short-course hepatic arterial
infusion chemotherapy with high-dose 5-uorouracil and cisplatin in
patients with advanced hepatocellular carcinoma. Cancer
2007;110:129 37
14 Kim BK, Park JY, Choi HJ, et al. Long-term clinical outcomes of hepatic
arterial infusion chemotherapy with cisplatin with or without
5-uorouracil in locally advanced hepatocellular carcinoma. J Cancer
Res Clin Oncol 2011;137:659 67
15 Seki H, Kimura M, Yoshimura N, et al. Hepatic arterial infusion
chemotherapy using percutaneous catheter placement with an
implantable port: assessment of factors affecting patency of the
hepatic artery. Clin Radiol 1999;54:221 7
16 Shindoh N, Ozaki Y, Kyogoku S, et al. Stabilization of a percutaneously
implanted port catheter system for hepatic artery chemotherapy
infusion. Cardiovasc Interv Radiol 1999;22:344 7
17 Chen Y, He X, Chen W, et al. Percutaneous implantation of a
port-catheter system using the left subclavian artery. Cardiovasc Intervent
Radiol 2000;23:2225
18 Venturini M, Angeli E, Salvioni M, et al. Complications after
percutaneous transaxillary implantation of a catheter for
intraarterial chemotherapy of liver tumors: clinical
relevance and management in 204 patients. Am J Roentgenol
2004;182:1417 26
19 Civelek AC, Sitzmann JV, Chin BB, et al. Misperfusion of the liver during
hepatic artery infusion chemotherapy: value of preoperative
angiography and postoperative pump scintigraphy. Am J Roentgenol
1993;160:865 70
20 Seki H, Kimura M, Kamura T, et al. Hepatic perfusion abnormalities
during treatment with hepatic arterial infusion chemotherapy: value of
CT arteriography using an implantable port system. J Comput Assist
Tomogr 1996;20:343 8
21 Morimoto M, Satake M, Sekiguchi R, et al. Optimal infection protocol for
CT evaluation during hepatic arterial infusion chemotherapy. Invest
Radiol 1999;34:744 50
22 Sone M, Kato K, Nakasato T, et al. Multislice CT angiography through an
implantable catheter and port system: early experience in detection
of vascular complications during hepatic arterial infusion chemotherapy.
J Comput Assist Tomogr 2002;26:515 9
23 Kapoor V, Brancatelli G, Federle MP, et al. Multidetector CT
arteriography with volumetric three-dimensional rendering to evaluate
patients with metastatic colorectal disease for placement of a oxuridine
infusion pump. Am J Roentgenol 2003;181:455 63
24 Sahani DV, Krishnamurthy SK, Kalva S, et al.
Multidetector-row computed tomography angiography for
planning intra-arterial chemotherapy pump placement in patients
with colorectal metastases to the liver. J Comput Assist Tomogr
2004;28:478 84
25 Lehmkuhl L, Denecke T, Warschewske G, et al. Multislice computed
tomographic angiography for preinterventional planning of port
placement for intra-arterial hepatic infusion chemotherapy. J Comput
Assist Tomogr 2007;31:66 71
26 Sone M, Kato K, Hirose A, et al. Impact of multislice CT angiography on
planning of radiological catheter placement for hepatic arterial infusion
chemotherapy. Cardiovasc Intervent Radiol 2008;31:91 7
27 Oken MM, Creech RH, Tormey DC, et al. Toxicity and response criteria
of the Eastern Cooperative Oncology Group. Am J Clin Oncol
1982;5:649 55
28 Tanaka T, Arai Y, Inaba Y, et al. Radiologic placement of side-hole
catheter with tip xation for hepatic arterial infusion chemotherapy.
J Vasc Interv Radiol 2003;14:63 8
29 Seki H, Ozaki T, Shiina M. Side-hole catheter placement for hepatic
arterial infusion chemotherapy in patients with liver metastases from
colorectal cancer: long-term treatment and survival benet.
Am J Roentgenol 2008;190:111 20
30 Seki H, Ozaki T, Takaki S, et al. Using slow-infusion MR
arteriography and an implantable port system to assess drug
distribution at hepatic arterial infusion chemotherapy. Am J Roentgenol
2003;180:681 6