Placement of a Long Tapered Side-Hole Catheter in the Hepatic Artery: Technical Advantages, Catheter Stability, and Arterial Patency

Transcription

1 Seki and Shiina Placement in Hepatic Arteries Interventional Radiology Original Research A C M E D E N T U R I C A L I M A G I N G AJR 2006; 187: X/06/ American Roentgen Ray Society Y Hiroshi Seki 1 Makoto Shiina Seki H, Shiina M O F Keywords: catheters, infusion chemotherapy, hepatic arteries, interventional radiology, implantable devices, liver DOI: /AJR Received May 1, 2005; accepted after revision September 21, Both authors: Department of Radiology, Niigata Cancer Center Hospital, , Kawagishi-cho, Niigata , Japan. Address correspondence to H. Seki (hseki@niigata-cc.jp). Placement of a Long Tapered Side-Hole in the Hepatic Artery: Technical Advantages, Stability, and Arterial Patency OBJECTIVE. The purpose of this study was to evaluate the technical advantages, safety, and efficacy of placing a catheter distally in the hepatic artery using a long tapered side-hole catheter with an implantable port for hepatic arterial infusion chemotherapy. SUBJECTS AND METHODS. Fifty patients with unresectable malignant liver tumors underwent radiologic implantation of catheter-port systems using the long tapered catheter placement method. A 2.7-French distal shaft of the catheter was inserted distally in the hepatic artery with its side hole located proximally, and a 5-French proximal shaft was placed in the aorta; the catheter tip was not fixed. Technical success, complications including catheter stability and hepatic artery patency, and tumor response were assessed and compared with the following two historical controls: 35 patients with a 5-French catheter inserted simply in the hepatic artery (conventional method), and 131 patients with a 5-French catheter, the tip of which was fixed in the gastroduodenal artery (the fixed-catheter-tip method). RESULTS. The technical success rate using the long tapered catheter placement method was 92% (46/50 patients), whereas the feasibility of the fixed-catheter-tip method was confined to 79% of historical controls (131/166 patients). Among patients in whom the gastroduodenal artery was present, a decreased frequency of gastroduodenal artery embolization was seen using the long tapered catheter placement method (39%; 17/44 patients) compared with the conventional method (p = ) and the fixed-catheter-tip method (p < ). Cumulative stability rates of the catheter (6 months, 94.9%; 1 year, 94.9%; 2 years, 86.2%) and cumulative patency rates of the hepatic artery (6 months, 89.9%; 1 year, 89.9%; 2 years, 83.5%) were significantly higher using the long tapered catheter placement method than using the conventional method (p = and p = , respectively) but were similar to those using the fixedcatheter-tip method. The time of hepatic tumor progression was significantly longer using the long tapered catheter placement method than using the conventional method (p = ) but was comparable to the time using the fixed-catheter-tip method. CONCLUSION. The long tapered catheter placement method should find wider application in hepatic arterial infusion chemotherapy because it is useful in preventing catheter dislodgment and hepatic artery occlusion. he liver is one of the most common sites of cancer metastases T that result in significant morbidity and mortality. For selected patients with isolated liver metastases, surgical resection can offer long-term survival. Unfortunately, however, most patients are not candidates for resection [1, 2]. Local tumor ablative therapies such as radiofrequency ablation and cryotherapy have shown promise as alternative minimally invasive treatments that are available to patients for whom surgical resection is contraindicated. Nevertheless, a high local recurrence rate has been reported [3, 4]. For unresectable liver metastases, hepatic arterial infusion chemotherapy has been widely used as a regional therapy, and a significant response rate is achieved despite poor control of extrahepatic progression [5 8]. On the other hand, for patients with liver metastases from colorectal cancer, current clinical trials involving systemic chemotherapy using irinotecan and oxaliplatin, combined with 5- fluorouracil-based regimens, have shown improved survival [9, 10]. However, response rates and times to hepatic progression are superior in patients receiving hepatic arterial infusion chemotherapy compared with those receiving systemic treatment [5 10]. In an 1312 AJR:187, November 2006

2 Placement in Hepatic Arteries attempt to prevent extrahepatic progression, combinations of hepatic arterial infusion with systemic chemotherapy are currently being investigated [11, 12]. In addition, recent studies have reported the effectiveness of hepatic arterial infusion chemotherapy for advanced hepatocellular carcinoma [13, 14]. Therefore, hepatic arterial infusion chemotherapy using catheter-port systems is an important therapy for unresectable liver malignancies. Radiologic implantation of catheter-port systems is an easier and safer procedure than surgical implantation, and it does not require administration of general anesthesia [15 17]. However, frequent catheter dislodgment and hepatic artery occlusion were noted as major limitations of radiologic implantation using conventional catheter placement, in which an end-hole catheter is simply inserted into the common or proper hepatic artery [17 21]. To resolve these problems, catheter placement using the fixed-catheter-tip method has been developed, in which the tip of a catheter is fixed in the gastroduodenal artery using coils or liquid glue [17 20]. However, several limitations still remain: This method is inadequate when the gastroduodenal artery is absent or too short; many arterial branches, including the gastroduodenal artery, must be embolized to avoid extrahepatic perfusion and to stabilize the catheter; and removing the catheter in cases of catheter dysfunction is troublesome because hepatic artery occlusion may result from migration of embolic agents. Also, thrombosis of critical organs, such as cerebral infarction, may occur because of thrombi clinging to the tip of the catheter [20]. We have developed a radiologic implantation method for catheter-port systems using a long tapered side-hole catheter in which a small-caliber (2.7-French), 20-cm-long distal shaft with a side hole is inserted distally into the hepatic artery, and a larger (5-French) proximal shaft is placed in the aorta (the long tapered catheter placement method). The purpose of our study was to prospectively evaluate the feasibility and safety of the long tapered catheter placement method, and to assess stability of the catheter and the patency of the hepatic artery during hepatic arterial infusion chemotherapy, compared with the conventional method and the fixed-catheter-tip method. Subjects and Methods Patients Between March 2002 and October 2004, 50 patients (39 men and 11 women; age range, years; mean age, 64 years) with malignant tumors of the liver were entered in this prospective study. Our institutional review board approved the study. Informed consent was obtained from each patient before the procedure was initiated. The inclusion criteria were as follows: liver tumors not suitable for surgical resection; a performance status of two or less according to the Eastern Cooperative Oncology Group classification [22]; a serum bilirubin level less than 3.0 mg/dl; and no prior transcatheter arterial chemoembolization for liver tumors. Several studies have indicated that transcatheter arterial chemoembolization is one of the causes of hepatic artery occlusion [18]. Therefore, we eliminated patients with prior transcatheter arterial chemoembolization for the liver from this study. Small extrahepatic disease confirmed by radiologic examinations or intraoperative findings was not considered an absolute contraindication if the liver was the predominant site of disease. Three patients had hepatocellular carcinoma, and 47 had liver metastases from colorectal cancer (n = 23), gastric cancer (n = 19), pancreatic cancer (n = 2), bile duct cancer (n = 1), esophageal cancer (n =1), and lung cancer (n = 1). Ten of these patients had recurrent liver tumors after resection of the liver: wedge resection in three, lateral or right posterior segmentectomy in two, left hepatic lobectomy in two, and both wedge resection and lateral segmentectomy in three patients. For comparison, 166 patients with catheterport systems using a 5-French indwelling catheter were studied retrospectively. These patients had been treated from approximately 5 years to 1 month before this study. These historical controls were divided into the following two groups: 35 patients received the catheter-port system using the conventional method, in which a catheter was simply inserted into the proper or common hepatic artery, and 131 patients were treated using the fixed-catheter-tip method, in which the catheter tip was fixed into the gastroduodenal artery using coils, with the side hole placed in the common hepatic artery. Technical Procedure All procedures were performed in the radiology suite using an angiography unit. Before catheter placement, angiography was performed via a transfemoral route for arterial road mapping and arterial redistribution for all patients [20, 21, 23]. When we encountered an aberrant hepatic artery, we decided to implant an indwelling catheter into the larger hepatic artery, and the smaller hepatic artery was embolized using coils (FPC35 Pt-Max and VortX, Boston Scientific; Tornado, Cook) to convert the multiple hepatic arteries into a single arterial blood supply. In addition, extrahepatic arterial branches arising from the hepatic artery were occluded, if necessary, using coils to avoid extrahepatic perfusion of chemotherapeutic drugs. placement was performed several days after angiography was performed. For premedication, 15 mg of pentazocine hydrochloride and 25 mg of hydroxyzine pamoate were given by an intramuscular injection. The procedure was done under local infiltration anesthesia using 1% procaine hydrochloride. In 47 of 50 patients, the left axillary artery was chosen as the access route using the following technique: A skin incision was made under the left clavicle approximately 4 cm long; the axillary artery was surgically exposed and its arterial branch, mainly the thoracoacromial artery, was cut down; a 5-French, 30-cm-long introducer sheath (Supersheath, Medikit) was inserted via this branch into the descending aorta. For the remaining three patients who had undergone prior angiography revealing that the celiac artery arose from the abdominal aorta at an acute angle, the inferior epigastric artery was used as the access route using the following procedure: A inch guidewire (Radifocus, Terumo) was inserted into the inferior epigastric artery with a contralateral transfemoral approach; the skin of the lower abdominal wall was incised and the inferior epigastric artery was exposed; the guidewire was pulled through a small incision in the inferior epigastric artery, and a 5-French, 10-cm-long introducer sheath (Supersheath) was advanced into the external iliac artery over this guidewire [17]. A long tapered-type catheter was used as an indwelling catheter (Anthron P-U, Toray). This was composed of a 2.7-French distal shaft 20 cm long and a 5-French proximal shaft that was 70 cm long. The catheter was made of polyurethane and its surface was coated with heparin. This catheter has been approved by the Ministry of Health, Labor, and Welfare of Japan for clinical application as an implantable medical device. In the long tapered catheter placement method used in this study, a slit-type side hole approximately 2 or 3 mm long and approximately 0.5 mm wide was made manually using small scissors in the 2.7-French distal shaft of the catheter. The 2.7- French distal shaft with the side hole was then placed at the abdominal aorta near the origin of the celiac artery trunk or the superior mesenteric artery and extending to a peripheral branch of the hepatic artery with no fixation of the catheter tip (Figs. 1 and 2), and the 5-French proximal shaft was placed between the aorta and the access artery. To decide the distance between the side hole and the tip of the catheter, we used the following technique: A 5-French angiographic catheter (Selecon catheter, Clinical Supply) was inserted into the hepatic artery through the access route; and a 3- AJR:187, November

3 Seki and Shiina French microcatheter (On The Road Infusion, Solution) was coaxially introduced into this catheter and advanced far into a peripheral hepatic artery using a inch, 300-cm-long guidewire (Rainbow, Piolax Medical Devices). Under fluoroscopic control, this guidewire was drawn out from the peripheral branch (the predetermined position A C of the catheter tip) to the proximal portion of the hepatic artery (the predetermined position of the side hole), and its length was measured. The 5- and 3- French angiographic catheters were exchanged for the indwelling catheter over the guidewire. After the introducer sheath was removed, the proximal portion of the indwelling catheter was fixed to the Fig year-old man with liver metastases from sigmoid colon cancer. A, Celiac arteriogram obtained before catheter placement shows right gastric artery (arrows) and accessory left gastric artery (arrowheads) arising from proper hepatic artery. B, Abdominal radiograph obtained just after catheter placement shows 2.7-French distal shaft of catheter in celiac artery and 5-French proximal shaft in aorta. tip is advanced distally into right hepatic artery and side hole of distal shaft is located in proper hepatic artery (arrow). C, Arteriogram using catheter-port system obtained just after implantation shows no extrahepatic perfusion and no embolization of gastroduodenal artery, and occlusion of right gastric artery (arrow) and accessory left gastric artery (arrowhead) using coils. arterial branch of the access route by ligation. Finally, the proximal end of the catheter was cut and connected to the port (Soph-A-Port, Sophysa), which was implanted in the subcutaneous space. Hepatic arterial infusion chemotherapy was started after an initial check of the catheter-port system using digital subtraction angiography, CT, B 1314 AJR:187, November 2006

4 Placement in Hepatic Arteries and hepatic arterial perfusion scintigraphy. The following chemotherapeutic drugs were used: 5- fluorouracil (1,000 mg/m 2 weekly in a 5-hour infusion) was administered to patients with liver metastases from colorectal cancer; 5-fluorouracil (330 mg/m 2 weekly in a 3-hour infusion), mitomycin C (2.7 mg/m 2 biweekly in a bolus infusion), and epirubicin (30 mg/m 2 every 4 weeks in a bolus infusion) for patients with liver metastases from gastric, pancreatic, and bile duct cancer; and 5-fluorouracil (1,000 mg/m 2 weekly in a 5-hour infusion) and cisplatin (10 mg weekly in a bolus infusion) for patients with hepatocellular carcinoma and liver metastases from esophageal and lung cancer. Among the 166 patients who were historical controls, the arterial access technique was performed in the same manner, and the 5-French catheter was made of polyurethane with a surface coating of heparin similar to that used in our study. In addition, the regimens of hepatic arterial infusion chemotherapy were similar to those used in this study. Follow-Up Patients were followed until the end of the hepatic arterial infusion chemotherapy. Abdominal A Fig year-old woman with liver metastases from rectal cancer who previously underwent left hepatic lobectomy. A, Arteriogram obtained before catheter placement shows replaced right hepatic artery arising from superior mesenteric artery. B, Arteriogram using catheter-port system was obtained just after implantation. Long tapered side-hole catheter was inserted into replaced right hepatic artery. Arterial branch of caudate lobe of liver arising from proper hepatic artery was occluded using coils (arrow). radiography, digital subtraction angiography via the implanted catheter, CT arteriography using the catheter-port system, or hepatic arterial perfusion scintigraphy was performed within 10 days after implantation of the catheter-port system and every 2 3 months thereafter to assess the position of the catheter, patency of the hepatic artery, and perfusion pattern in the liver. Digital subtraction angiography was performed during injection of the contrast medium through the catheter-port system. CT arteriography using the catheter-port system was performed through the entire liver, in which the contrast medium of 100 mg I/mL of iopamidol (Iopamiron, Schering) was injected via the implanted port at a rate of 0.7 ml/s, and data acquisition began 20 seconds after initiation of injection [17, 24]. Hepatic artery perfusion scintigraphy was performed using 370 MBq of technetium-99m macroaggregated albumin in 10 ml of saline, which was infused for 1 hour through the implanted port at a rate of 10 ml/h. In general, hepatic arterial perfusion scintigraphy was performed during the initial perfusion study, and CT arteriography was performed at the initial and follow-up evaluations. These imaging procedures were also performed when patients complained of any symptoms related to hepatic artery infusion with chemotherapeutic drugs. In addition, a whole-body CT examination was performed before treatment and every 3 months thereafter to evaluate treatment responses. Evaluation of Technical Outcomes We evaluated the technical success using the long tapered catheter placement method. In addition, we recorded what kind of artery was occluded when coils were used to avoid extrahepatic perfusion of infused drugs because of the arterial anatomy. Evaluation of catheter stability, hepatic artery patency, other complications, catheter dysfunction, and perfusion pattern During the follow-up period, stability of the catheter and patency of the hepatic artery were assessed. They were measured from the date of catheter placement until the date of catheter dislodgment or hepatic artery occlusion, or until the end of the hepatic arterial infusion chemotherapy. In addition, other complications such as catheter dysfunction and perfusion abnormality were evaluated. Evaluation of tumor responses The objective tumor response was assessed using an abdominal CT examination. Responsive cases were defined as those with a 50% or more decrease in the sum of the products of the perpendicular diameters of B AJR:187, November

5 Seki and Shiina TABLE 1: Hepatic Artery Receiving the Inserted Long Tapered Side-Hole and Position of the Hepatic Artery Receiving Inserted Position No. Hepatic Artery Arising from Celiac Artery (n = 38) Replaced RHA Arising from SMA (n = 5) Replaced PHA Arising from SMA (n = 2) Replaced CHA Arising from SMA (n = 1) Tip position Right anterosuperior segmental branch Right posteroinferior segmental branch Right anteroinferior segmental branch Right posterosuperior segmental branch Left anterolateral segmental branch Position of side hole PHA CHA Proximal portion of replaced RHA Note RHA = right hepatic artery, PHA = proper hepatic artery, CHA = common hepatic artery, SMA = superior mesenteric artery. TABLE 2: Embolization of the Gastroduodenal Artery (GDA) to Avoid Extrahepatic Perfusion Presence or Absence of GDA, by Placement Method Long Tapered Placement Method Conventional Method Fixed--Tip Method Completion of Embolization Present (n = 44) Absent (n = 2) Present (n = 24) Absent (n = 11) Present (n = 131) Absent (n = 0) Yes No the lesions. Nonresponsive cases were described as a 50% or less decrease, or any increase, in the sum of the areas of the indicator lesions. Statistical Analysis Comparisons of patient characteristics for the three groups treated with the long tapered catheter placement method, the conventional method, and the fixed-catheter-tip method were performed using the chi-square test and Fisher s exact probability test. In addition, completion of embolization of the gastroduodenal artery and the objective tumor response were compared among the three groups using these tests. The cumulative stability rate of the catheter, the patency rate of the hepatic artery, and the time to hepatic progression were calculated using the Kaplan-Meier method, and the resultant curves were compared using the log-rank test. Statistical significance was established with a p value of less than Results Technical Success The catheter-port system was successfully implanted using the long tapered catheter placement method in 46 (92%) of 50 patients, who experienced no complications related to the procedure. The group in which the procedure failed consisted of three patients in whom the caliber of the proper hepatic artery was too small ( 2 mm in diameter) to introduce the catheter distally. For these patients, we judged that the hepatic artery might be occluded if the catheter was implanted into it. Therefore, the fixed-catheter-tip method was used instead. For the one remaining patient, arterial stenosis was found at the junction between the common and the proper hepatic arteries, probably as a result of postoperative changes after a prior pancreatoduodenectomy. Therefore, a modified fixed-catheter-tip method was selected in which the side hole of the catheter was placed in the celiac artery trunk and the catheter tip was fixed into the splenic artery using coils. The hepatic artery into which the long tapered side-hole catheter was inserted, and the position of the catheter, are presented in Table 1. The long tapered side-hole catheter was inserted into the hepatic artery arising from the celiac artery in 38 patients (Fig. 1) and into the aberrant hepatic artery arising from the superior mesenteric artery in eight patients (Fig. 2). Among these patients, the gastroduodenal artery was absent because of pervious surgery in two patients. The catheter could be implanted using the fixed catheter-tip method in 131 (79%) of 166 historical patients. The conventional method was used for the remaining 35 patients, in whom the gastroduodenal artery was absent or was too short to fix the catheter tip because of previous surgery. Embolization of Extrahepatic Arterial Branches The rate of completion of embolization of the gastroduodenal artery to avoid extrahepatic perfusion is listed in Table 2. Among patients in whom the gastroduodenal artery was present, the gastroduodenal artery was embolized using coils in 17 (39%) of 44 patients because of the following causes: For 14 patients the side hole of the catheter had to be placed in the common hepatic artery or the proper hepatic artery immediately near the origin of the gastroduodenal artery because of a short proper hepatic artery; for the remaining three patients, the gastroduodenal artery was occluded when the origin of the proper hepatic artery was embolized using coils to convert the hepatic artery blood flow into the replaced right hepatic artery arising from the superior mesenteric artery AJR:187, November 2006

6 Placement in Hepatic Arteries TABLE 3: Patient Characteristics and Placement Methods For the remaining 27 patients, however, embolization of the gastroduodenal artery was not required for the following reasons: The side hole of the catheter could be placed in the proper hepatic artery at a distance from the origin of the gastroduodenal artery in 23 patients (Fig. 1), and the catheter was inserted in the replaced right or proper hepatic artery arising from the superior mesenteric artery in four patients (Fig. 2). On the other hand, using the conventional method, the gastroduodenal artery was embolized using coils Long Tapered Placement (n = 46) Method Conventional (n = 35) Fixed--Tip (n = 131) Characteristics Age (y) 60 / > / / / 80 Sex Male / female 37 / 9 26 / 9 92 / 39 Primary or metastatic liver tumor Primary / metastatic 3 / 43 2 / 33 3 / 128 % Liver involvement 60% / > 60% 31 / / / 47 Prior liver resection Yes / no 10 / 36 6 / / 96 Aberrant hepatic artery Yes / no 13 / 33 8 / / 100 Access route Axillary artery / inferior epigastric artery 43 / 3 28 / / 22 TABLE 4: Dislodgment Method Dislocation Long Tapered Placement Conventional Fixed--Tip Withdrawal of catheter side hole from PHA to CHA Distal portion to proximal region of CHA CHA to celiac arterial trunk Withdrawal of catheter-tip in Splenic artery Celiac arterial trunk Abdominal aorta Advancement of catheter tip in right hepatic artery Note PHA = proper hepatic artery, CHA = common hepatic artery. in 17 (71%) of 24 patients because the catheter had to be placed in the common hepatic artery as a result of the difficulty of stable catheter placement in the proper hepatic artery. Using the fixed-catheter-tip method, the gastroduodenal artery was occluded using coils for all patients (100%). The frequency of embolization of the gastroduodenal artery was significantly lower using the long tapered catheter placement method than using the conventional method (p = ) or the fixed-catheter-tip method (p < ). The right gastric artery, which originated from the proper or left hepatic artery, if present, was occluded using coils to prevent gastric toxicity resulting from any catheter placement method (Fig. 1). A posterosuperior pancreaticoduodenal artery arising from the proper hepatic artery and an accessory left gastric artery arising from the left or proper hepatic artery were found on angiography in 4 and 3 patients, respectively, who were treated with the long tapered catheter placement method; in 3 and 1 patient, respectively, treated with the conventional method; and in 6 and 13 patients, respectively, treated with the fixed-catheter-tip method. All of these were occluded using coils (Fig. 1). In addition, the dorsal pancreatic artery arising from the common hepatic artery was embolized using coils in 3 patients who were treated with the conventional method, and in 12 patients treated with the fixed-catheter-tip method. Stability of the and Patency of the Hepatic Artery After implantation of catheter-port systems using the long tapered catheter placement method, 46 patients were followed up for a mean period of 283 days (range, days). Among the historical controls, the mean follow-up period was 255 days (range, 17 1,260 days) for 35 patients treated using the conventional method and 259 days (range, 16 1,162 days) for 131 patients treated using the fixed-catheter-tip method. Patient characteristics are shown in Table 3. The variables listed show no significant differences between patients among the three arms of the catheter placement method study. dislodgment data are listed in Table 4. Dislocation of the catheter was found using follow-up digital subtraction angiography through the catheter-port system in three (6.5%) of 46 patients treated with the long tapered catheter placement method. No correlation was seen between the access route and catheter dislodgment. The patients had no clinical symptoms related to dislocation of the catheter. They did not require repositioning of the catheter because the gastroduodenal artery had been occluded using coils at the implantation of the catheter-port system in two patients, and the gastroduodenal artery had hepatopetal blood flow in the third patient. Among historical controls treated using the conventional method, catheter dislodgment was seen in eight patients. Of these patients, the catheter was repositioned in four patients and systemic chemotherapy was substituted, AJR:187, November

7 Seki and Shiina Cumulative Stability of (%) Time (mo) Fig. 3 Graph shows Kaplan-Meier analysis for cumulative stability of catheter according to catheter placement method. = long tapered catheter placement method, censored cases; = conventional method, censored cases; X = fixedcatheter-tip method, censored cases. Stability rates in patients treated with long tapered catheter placement method were significantly higher than those for patients treated with conventional method (p = using log-rank test) but were similar to rates for those treated with fixed-catheter-tip method. Cumulative Patency of Hepatic Artery (%) resulting in progressive disease in these four patients. In the other control group treated with the fixed-catheter-tip method, catheter dislocation was seen in four patients. In these patients, the catheter was not removed and hepatic arterial infusion chemotherapy was continued thereafter. For two of these patients, however, dosage reduction of infused drugs was required. The cumulative stability rates of the catheters used in the long tapered catheter placement method were 94.9%, 94.9%, and 86.2% at 6 months, 1 year, and 2 years, respectively. These were significantly higher than those for the conventional method (p = ) but were similar to those using the fixed-catheter-tip method; in the latter case, no statistically significant difference was seen (p = ) (Fig. 3). Hepatic artery occlusion was observed in four (8.7%) of 46 patients treated with the long tapered catheter placement method. For three of these patients, hepatic artery occlusion was detected using follow-up CT arteriography and digital subtraction angiography through the catheter-port system, and these patients had no clinical symptoms. The one remaining patient complained of epigastric discomfort during the infusion of chemotherapeutic drugs, and hepatic artery occlusion was found using digital subtraction angiography through the catheter-port system. The sites of arterial occlusion were the proper hepatic artery, the right hepatic artery, the right anterior segmental branch, and the right posterior segmental branch. In these patients, the tip of the catheter had been inserted into a segmental branch of the right hepatic artery. Although the patients were treated with systemic chemotherapy, their incidence of liver tumors increased thereafter. Among the historical controls, hepatic artery occlusion was seen in 11 patients treated with the conventional method and in 16 patients treated with the fixed-catheter-tip method. All received systemic chemotherapy thereafter. Among these patients, progressive disease was subsequently seen in 23 patients, whereas the substituted treatment was effective in four patients. For patients treated with the long tapered catheter placement method, cumulative patency rates of the hepatic artery at 6 months, 1 year, and 2 years were 89.9%, 89.9%, and 83.5%, respectively. Statistically significant better patency of the hepatic artery occurred in patients treated with the long tapered catheter placement method than in those treated with the conventional method (p = ), whereas no statistically significant difference was seen compared with results for the fixed-catheter-tip method (p = ) (Fig. 4) Time (mo) Fig. 4 Graph shows Kaplan-Meier analysis to determine cumulative patency of hepatic artery according to catheter placement method. = long tapered catheter placement method, censored cases; = conventional method, censored cases; X = fixed-catheter-tip method, censored cases. Patency rates for patients treated with long tapered catheter placement method were significantly higher than those for patients treated with conventional method (p = using log-rank test) but were not statistically different from those treated with fixed-catheter-tip method. Other Complications, Dysfunction, and Perfusion Pattern Complications, catheter dysfunctions, and perfusion abnormalities are summarized in Table 5. For the long tapered catheter placement method, except for hepatic artery occlusion and catheter dislodgment, major complications did not occur. Occlusion of the catheter was not seen and the side hole of the catheter was patent during the follow-up period. In two patients, a collateral blood supply to the liver developed. Collateral vessels were confirmed using angiography via a transfemoral approach, and they were embolized to correct intrahepatic drug distribution. Initial perfusion studies using the catheterport system showed a sufficient distribution through the entire liver in all 46 patients treated with the long tapered catheter placement method. Angiography was performed just after catheter implantation and showed that the contrast medium flowed from both the side hole and the end hole of the implanted catheter. Follow-up digital subtraction angiography through the implanted catheter showed that the outflow from the side hole of the catheter became predominant compared with that from its end hole. CT arteriography using the catheter-port system, however, showed that hypoperfusion in the liver where the catheter was inserted distally was not seen in any patients. Tumor Response and Time to Hepatic Progression Tumor response and the time to hepatic progression are presented in Table 6. No statistically significant difference was seen in the objective tumor response rate among the three arms of the catheter placement method 1318 AJR:187, November 2006

8 Placement in Hepatic Arteries TABLE 5: Complications, Dysfunctions, and Perfusion Abnormalities Long Tapered Complications Placement Conventional Fixed--Tip Procedure-related complications Migration of coils Hematoma around port Complications during chemotherapy Hepatic arterial occlusion Gastroduodenal ulcer Cholangitis Cerebral infarction Thrombosis of axillary artery dysfunction dislodgment Occlusion of catheter kinking near port Perfusion abnormality Development of collateral blood supply to liver study. However, the time to hepatic progression was significantly longer for patients treated with the long tapered catheter placement method than for those treated with the conventional method (p = ) but was comparable to the time associated with the fixed-catheter-tip method (p = ). Method TABLE 6: Tumor Response and Time to Hepatic Progression Method Long Tapered Response Placement (n = 46) Conventional (n = 35) Fixed--Tip (n = 131) Tumor response Number Response rate (%) Median time to hepatic progression (mo) Discussion Previous studies have described catheter dislodgment and hepatic artery occlusion as major complications that can accompany radiologic implantation using the conventional method [18 21]. The high incidence of catheter dislodgment may be related to the indwelling catheter s being too short in the hepatic artery [15, 21]. Mechanical trauma caused by the catheter tip on the arterial wall is considered by some to be a major cause of artery occlusion [17 21]. Recently, radiologic implantation using a fixed-catheter-tip method has been designed by several investigators in an attempt to resolve these problems [17 20]. However, this method is technically complicated, and it cannot be applied in patients in whom the gastroduodenal artery is absent or is too short to insert the catheter. The long tapered catheter placement method used in this study does not require fixation of the catheter tip. Also, if necessary, the catheter can be removed easily and safely. In our study, the long tapered catheter placement method could be used in 92% of patients, regardless of the presence or absence of the gastroduodenal artery, whereas the fixed-catheter-tip method was feasible in only 79% of historical controls. In addition, compared with the conventional method and the fixedcatheter-tip method, the long tapered catheter placement method had a decreased frequency of embolizing the gastroduodenal artery because the side hole of the catheter could be placed stably in the proper hepatic artery in many cases. The stability rate of the catheter and the patency of the hepatic artery were significantly higher in patients treated with the long tapered catheter placement method than in those treated with the conventional method, and were comparable to rates for patients treated with the fixed-catheter-tip method. In the long tapered catheter placement method, catheter insertion distally in the hepatic artery contributed to a reduction in mobility of the catheter, resulting in a decreased incidence of catheter dislocation and a reduced mechanical stimulation of the arterial wall. In addition, using this method, the distal shaft of the catheter placed in the hepatic artery had a small caliber (2.7 French) and was made of a soft material (polyurethane) coated with heparin. These characteristics made it less likely to generate hepatic artery thrombosis. dislodgment and hepatic artery occlusion may decrease the effectiveness of hepatic arterial infusion chemotherapy because of temporary or permanent interruption of the intraarterial infusion of drugs. In our study, the time to hepatic progression was significantly longer for patients treated with the long tapered catheter placement method than for those treated with the conventional method, and was comparable to the time required for those treated with the fixed-catheter-tip method. Although the control groups in this study were historical, these results indicate that the long tapered catheter placement method is comparable to the fixed-catheter-tip method with respect to preventing catheter dislodgment and hepatic artery occlusion. In our study, the fixed-catheter-tip method was used instead of the long tapered catheter placement method for a few patients in whom the caliber of the proper hepatic artery was too small to insert the catheter distally. Therefore, the fixed-catheter-tip method will still be applicable as an alternative catheter placement method in these special circumstances. In conclusion, the long tapered catheter placement method should find wide application because it is an easy interventional procedure and it will help prevent catheter dislodgment and hepatic artery occlusion during hepatic arterial infusion chemotherapy. We believe that this method can be used for radiologic implantation of catheter-port systems instead of the current standard, the fixed-cath- AJR:187, November

9 Seki and Shiina eter-tip method, although alternative methods still must be used in some cases. References 1. Jamison RL, Donohue JH, Nagorney DM, Rosen CB, Harmsen WS, Ilstrup DM. Hepatic resection for metastatic colorectal cancer results in cure for some patients. Arch Surg 1997; 132: Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999; 230: Solbiati L, Livraghi T, Goldberg SN, et al. Percutaneous radio-frequency ablation of hepatic metastases from colorectal cancer: long-term results in 117 patients. Radiology 2001; 221: Lencioni R, Crocetti L, Cioni D, Della Pina C, Bartolozzi C. Percutaneous radiofrequency ablation of hepatic colorectal metastases: technique, indications, results, and new promises. Invest Radiol 2004; 39: Lorenz M, Muller HH. Randomized, multicenter trial of fluorouracil plus leucovorin administered either via hepatic arterial or intravenous infusion versus fluorodeoxyuridine administered via hepatic arterial infusion in patients with nonresectable liver metastases from colorectal carcinoma. J Clin Oncol 2000; 18: Kemeny N, Conti JA, Cohen A, et al. Phase II study of hepatic arterial floxuridine, leucovorin, and dexamethasone for unresectable liver metastases from colorectal carcinoma. J Clin Oncol 1994; 12: Kemeny N, Neidwiecki D, Hollis DR, et al. Hepatic arterial infusion versus systemic therapy for hepatic metastases from colorectal cancer: a randomized trial of efficacy, quality of life, and molecular markers (CALGB 9481). J Clin Oncol 2006; 24: Cohen AD, Kemeny NE. An update on hepatic arterial infusion chemotherapy for colorectal cancer. Oncologist 2003; 8: Maindrault-Goebel F, de Gramont A, Louvet C, et al. High-dose intensity oxaliplatin added to the simplified bimonthly leucovorin and 5-fluorouracil regimen as second-line therapy for metastatic colorectal cancer (FOLFOX 7). Eur J Cancer 2001; 37: Rothenberg ML, Oza AM, Bigelow RH, et al. Superiority of oxaliplatin and fluorouracil-leucovorin compared with either therapy alone in patients with progressive colorectal cancer after irinotecan and fluorouracil-leucovorin: interim results of a phase III trial. J Clin Oncol 2003; 21: Kemeny N, Jarnagin W, Gonen M, et al. Phase I/II study of hepatic arterial therapy with floxuridine and dexamethasone in combination with intravenous irinotecan as adjuvant treatment after resection of hepatic metastases from colorectal cancer. J Clin Oncol 2003; 21: Zelek L, Bugat R, Cherqui D, et al. Multimodal therapy with intravenous biweekly leucovorin, 5-fluorouracil and irinotecan combined with hepatic arterial infusion pirarubicin in non-resectable hepatic metastases from colorectal cancer (a European Association for Research in Oncology trial). Ann Oncol 2003; 14: Ando E, Tanaka M, Yamashita F, et al. Hepatic arterial infusion chemotherapy for advanced hepatocellular carcinoma with portal vein tumor thrombosis: analysis of 48 cases. Cancer 2002; 95: Tanioka H, Tsuji A, Morita S, et al. Combination chemotherapy with continuous 5-fluorouracil and low-dose cisplatin infusion for advanced hepatocellular carcinoma. Anticancer Res 2003; 23: 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: Herrmann KA, Waggershauser T, Sittek H, Reiser MF. Liver intraarterial chemotherapy: use of the femoral artery for percutaneous implantation of catheter-port systems. Radiology 2000; 215: Arai Y, Inaba Y, Takeuchi Y. Interventional techniques for hepatic arterial infusion chemotherapy. In: Castaneda-Zuniga WR, ed. Interventional radiology, 3rd ed. Baltimore, MD: Williams & Wilkins, 1997: Seki H, Kimura M, Yoshimura N, Yamamoto S, Ozaki T, Sakai K. 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: Irie T. Intraarterial chemotherapy of liver metastases: implantation of a microcatheter-port system with use of modified fixed catheter tip technique. J Vasc Interv Radiol 2001; 12: Tanaka T, Arai Y, Inaba Y, et al. Radiologic placement of side-hole catheter with tip fixation for hepatic arterial infusion chemotherapy. J Vasc Interv Radiol 2003; 14: 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. AJR 2004; 182: Oken MM, Creech RH, Tormey DC, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol 1982; 5: Yamagami T, Kato T, Iida S, Tanaka O, Nishimura T. Value of transcatheter arterial embolization with coils and n-butyl cyanoacrylate for long-term hepatic arterial infusion chemotherapy. Radiology 2004; 230: Seki H, Kimura M, Kamura T, Miura T, Yoshimura N, Sakai K. 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: AJR:187, November 2006