 |
       |
|
|
TRANSCATHETER ARTERIAL CHEMOEMBOLIZATION OF HEPATOCELLULAR CARCINOMA
JIN WOOK CHUNG, JAE HYUNG PARK
Department of Radiology, Seoul National University College of Medicine, 28 Yongon-dong, Chongno-gu, Seoul, 110 744, Korea
INTRODUCTION
Hepatocellular carcinoma (HCC) is a highly malignant tumour of liver cell origin showing poor prognosis. Definitive surgical intervention is not feasible in most cases because of extreme tumour extension, multiplicity of tumour foci, and associated advanced liver cirrhosis at the time of diagnosis. Only 15-30 % of patients with HCCs are eligible for resection and, even after curative resection, the postoperative recurrence is indeed common. Because of these reasons, transcatheter approach has been aggressively tried to treat HCCs. Transcatheter arterial embolization for HCC was first reported by Doyon et al.[1] and chemoembolization using gelfoam and anticancer drugs was reported by Yamada et al.[2] In the early 1980s, iodized oil (Lipiodol; Andre Guerbet, Aulnay-sous-Bois, France), a lymphangiographic dye, was found to remain selectively in the neovasculature and extravascular spaces of HCC when it was injected into the hepatic artery.[3,4] Thereafter, transcatheter oily chemoembolization (TOCE) with an iodized oil mixed with different anticancer agents or, more often, with the emulsion of iodized oil and anticancer drugs followed by gelfoam embolization, has been increasingly used as an effective means of palliation for unresectable or postoperatively recurrent tumours and as an alternative to surgery even for resectable ones.
PRINCIPLES
The liver receives a dual blood supply from the hepatic artery and the portal vein. Approximately one-third of the normal hepatic blood flow comes from the hepatic artery and the other two-thirds from the portal vein. About half of the oxygen required by the liver is supplied by the portal vein. Instead, HCC is almost exclusively supplied by the hepatic artery, which can be easily recognized on CT hepatic arteriography and CT arterioportography. In the past, hepatic artery ligation has been tried to induce ischaemic necrosis of the tumour.[5] But, its effect of tumour regression and symptomatic improvement was temporary because proximal occlusion of the hepatic artery induces rapid development of collateral circulation to the peripheral hepatic arteries. Peripheral hepatic artery embolization can minimize the development of collateral circulation and induce selective ischaemic necrosis of HCC without serious ischaemic injury of normal liver parenchyma. For this purpose, absorbable gelatin sponge particles cut into pieces 1 to 2 mm in size are mostly frequently used. Embolization of the hepatic artery with gelatin sponge particles alone does not cause serious hepatic damage in experimental animals and humans with good hepatic functional reserve. Too small-sized embolic particles such as silicone rubber or gelfoam powder (particle size of 80 to 200 m) can induce liver infarction and bile duct necrosis.
TOCE is the combination of hepatic artery embolization and regional chemotherapy with use of an iodized oil not only as an embolic material but also as a carrier of chemotherapeutic agents. Probably because of the haemodynamic difference between hypervascular hepatic tumours and the liver parenchyma, the iodized oil injected into the hepatic artery rather selectively retains in hypervascular tumours including HCC. If a stable mixture of an iodized oil and an anticancer drug is infused into the hepatic artery, the anticancer drug can be locally delivered to HCC together with the iodized oil as a carrier. Nakamura et al.[6] demonstrated that the drug was released slowly from the emulsion in patients who underwent hepatectomy. In contrast, in the normal liver parenchyma, the iodized oil injected into the hepatic artery usually does not occlude the hepatic artery and accumulates in the peripheral portal vein through multiple arterioportal communications and subsequently passes through the sinusoids into the systemic circulation.[7] If a sufficient amount of iodized oil is injected to fill in the peripheral portal vein around the tumour (Fig. 1), TOCE with the combination of the iodized oil and gelatin sponge particles has the effect of combined arterial and portal blockage. Although the encapsulated nodular HCC is totally supplied by the hepatic artery, well-differentiated or early HCC and extracapsular infiltrating edge of advanced HCC are partly supplied by the portal vein. The shunted mixture of iodized oil and anticancer drugs into the portal vein may work to treat the tumour fraction partly supplied by the portal vein.
In clinical practice, the administration of the iodized oil mixed with anticancer drugs is followed by blockade of the hepatic arterial flow with gelatin sponge particles if the hepatic arterial flow does not slow down after iodized oil injection. Gelatin sponge embolization is needed to induce ischaemic necrosis of bulky tumours and to prevent untoward portal or systemic embolization of the iodized oil. Furthermore, in addition to the ischaemic effect of its own, gelatin sponge embolization contributes to prolonged retention of the iodized oil and the maintenance of a high local concentration of chemotherapeutic agents. As expected, findings of several clinical studies have proved that TOCE performed with the combination of iodized oil mixed with anticancer drugs and gelatin sponge particles is superior to hepatic artery embolization performed with iodized or gelatin sponge particles separately mixed with anticancer drugs.
PROCEDURES
First of all, underlying hepatic functional reserve is evaluated with laboratory tests and the presence or absence of ascites and hepatic encephalopathy. Through imaging studies, the size and extent of the tumour, the tumour growth pattern (expansile versus infiltrative), macroscopic angioinvasion into the hepatic and portal vein are evaluated. For successful segmental embolization, accurate segmental localization of the tumour is crucial and cross-sectional images are useful for that purpose. As complete angiography, all of hepatic arteries should be adequately opacified and all of feeding arteries should be carefully identified. Considering common anatomic variations of hepatic arteries, we routinely perform coeliac and superior mesenteric arteriography and evaluate the aberrant left hepatic artery from the left gastric artery (Fig. 2). Frequently, additional views in different angle and magnification and superselective injection with use of microcatheters are necessary to identify small feeding arteries. In order to avoid non-target embolization, it is important to recognize the origin of the cystic artery and the right gastric artery and the existence of long falciform artery and the accessory left gastric artery originating from the left hepatic artery (Fig. 3). We prospectively performed selective left hepatic arteriography in consecutive 100 patients and found long falciform artery and accessory left gastric artery arising from the left hepatic artery in 15 and 21 % of patients, respectively.
supplied by the right inferior phrenic artery. d) Selective arteriography of the right internal
mammary artery taken on the second session of treatment reveals focal tumour strain
The treatment protocol is individualized according to the hepatic functional reserve, tumour extent, and major portal vein invasion. The best way to maximize the treatment effect and to minimize procedure-related complications is to perform superselective chemoembolization of all tumour feeders. We can safely and effectively perform segmental or subsegmental chemoembolization for patients with small HCC and Child class C. However, the coexistence of extensive parenchymal tumour, major portal vein invasion, and poor liver function of Child class B or C is a definite contraindication of chemoembolization. We performed TOCE initially with infusion of iodized oil and doxorubicin hydrochloride emulsion. Every 10 mg of doxorubicin hydrochloride was dissolved in 0.5 mL of the water soluble contrast medium iopamidol; iodized oil and dissolved doxorubicin hydrochloride then were drawn separately into syringes interconnected with a three-way stopcock and emulsified by means of vigorous pushing of each syringes in alternation. The dose of iodized oil and doxorubicin hydrochloride depended on the size and vascularity of the tumour. Our current upper limit of iodized oil and doxorubicin hydrochloride is 15 mL and 50 mg, respectively. For a small tumour, we could administer enough iodized oil to saturate the entire tumour neovasculature. The end point for the emulsion administration was stasis in tumour feeding arteries with or without appearance of iodized oil in portal vein branches (Fig 1). When initial blockade of the hepatic artery was insufficient because of the large size of the mass or associated arterioportal shunting and the patient had Child class A disease, the emulsion and gelatin sponge particles, which were 1-2 mm in diameter and soaked with crystalline mitomycin, were alternately administered by use of a "sandwich technique". The dose of mitomycin ranged from 2 to 6 mg. In the presence of prominent arterioportal shunting, only a small amount of the emulsion was injected into the hepatic artery before gelatin sponge particle embolization. After adequate blockage of the shunt with gelatin sponge particle embolization, we injected an additional small amount of the emulsion. However, in the absence of arterioportal shunting, more than half of the emulsion prepared was injected into the hepatic artery before the switch to the sandwich technique.
RESULTS
There are many papers providing the overall survival rates after TOCE. In our series of 1067 patients, the 1-, 3-, 5-year overall survival rates were 77%, 46%, 23%, respectively.[8] The survival time of the patients with HCC largely depends on the presence or absence of prognostic factors irrespective of treatment modalities. The important prognostic factors are 1) hepatic functional reserve of the patient; 2) the size, number, and growth pattern of the tumour; 3) the presence or absence of angioinvasion, especially the major portal vein invasion. It is meaningless to compare our survival rates with those of different studies because the study population of every report shows different profile of prognostic factors.
There is no doubt about the effectiveness of segmental or subsegmental chemoembolization. Matsui et al.[9] reported complete necrosis in about 70% of 100 small hypervascular HCCs under 4 cm in diameter treated by subsegmental embolization with Lipiodol and ethanol. One year, 3 year, 5 year survival rates were 100%, 73%, and 53%, respectively, and 1 year and 5 year local recurrence rates following single treatment session were 18% and 33%, respectively. These results of TOCE can be favourably compared with the results of surgical resection and liver transplantation.[10] Recently, we have performed a prospective cohort study to evaluate therapeutic efficacy of TOCE as compared with hepatic resection for operable HCC. TOCE was as effective as hepatic resection in the subpopulations with tumour stage of UICC T3N0M0 and adequate liver function, and even with UICC T1-2N0M0 HCC when an iodized oil was compactly retained in the tumour.[11]
We evaluated the therapeutic efficacy and complications of TOCE in 110 patients with HCC invading main portal vein or its first order branches. The cumulative survival rates were poor and they were 30% (1 year), 18% (2 years), and 9% (3 years). However, 22 of 33 patients with a parenchymal tumour limited to one or two segments of a hepatic lobe, 22 had complete or partial remission, with a median survival time of 22 months; this survival time was significantly longer than that (5 months) for 77 patients with a more extensive tumour (p < 0.0001) (Fig. 4). Hepatic insufficiency developed in 10 patients, and three of them died within one month after chemoembolization. All 10 patients had an extensive parenchymal tumour involving more than two hepatic segments, and four had impaired hepatic functional reserve of Child's class B. So, we concluded that, when a tumour is limited in extent and hepatic function is preserved, TOCE is effective and safe for the palliation of HCC with major portal vein invasion, however, when a parenchymal tumour is extensive, chemoembolization is associated with a poor response and a risk of hepatic failure.[12]
In the far advanced stage of HCC, there is still a controversy about the effectiveness of TOCE. The controversy comes from the lack of standard treatment protocol and few prospective randomized studies. Prospective randomized studies performed in Western countries failed to provide any evidence of a survival benefit in patients treated with TOCE when compared to untreated patients.[13] Technical advance for selective catheterization of feeding arteries and collateral vessels and growing knowledge about the mechanisms, results, and limitations of TOCE can improve the effectiveness of the treatment and prolong the survival period of the patients and reduce TOCE-related complications.
COMPLICATIONS
TOCE is associated with diverse complications. However, the patients with procedure-related complications usually have predisposing factors or technical problems. Important predisposing factors are major portal vein obstruction, compromised hepatic functional reserve, biliary obstruction, previous biliary surgery, excessive amount of iodized oil, hepatic arterial occlusion after repeated TOCE, and nonselective embolization.[14] Recognition of predisposing factor before the procedure and selective embolization with adequate amount of chemoembolic material and careful post-procedure monitoring of the patients can prevent major complications in most patients.
The most frequent and serious complication of TOCE is hepatic insufficiency. Although the iodized oil is selectively retained in tumour neovasculature, TOCE also adversely affects nontumorous liver parenchyma. The normal liver of experimental animals tolerates even a large amount of iodized oil (0.5-2.0 mL/kg) well. However, treatment with the iodized oil alone is also ineffective in hepatic tumours. The addition of anticancer drugs to the iodized oil increases the therapeutic effect on malignant tumours with some hepatic damage, which includes small area of hepatic infarction.[15,16] Hepatic artery embolization with the combination of the iodized oil and gelatin sponge particles has the effect of combined arterial and portal blockage. Consequently, with the better therapeutic effects, hepatic artery embolization with the combination has a higher risk of hepatic necrosis and atrophy of nontumorous liver parenchyma than does embolization with iodized oil or gelatin sponge particles alone. Portal vein obstruction is a well-known risk factor for hepatic failure and infarction after hepatic artery embolization. In patients with major portal vein obstruction, TOCE should be performed with a reduced amount of chemoembolic agents and selective administration of them into tumour feeding arteries. Especially when there are other predisposing factors to hepatic insufficiency, TOCE seems to be contraindicated because of the risk of acute hepatic failure. Cirrhosis and chronic hepatitis frequently compromise the liver with HCC. A safe dose of iodized oil to the compromised liver has not been determined as yet. It is generally recommended that, in patients with Child-Pugh class B or C, the amount of the iodized oil used should be limited to less than 5-10 mL according to the severity of hepatic decompensation and the iodized oil should be injected to tumour feeding arteries as selective as possible.
In an experimental study, the liver with biliary obstruction was susceptible to hepatic infarction, which can be explained by means of the decreased portal venous inflow that accompanies biliary obstruction. Acute hepatic failure and sepsis can develop after TOCE for HCC with biliary obstruction.[14] Usually, the patients with biliary obstruction have extensive parenchymal tumour. In that case, adequate decompression of the biliary system should be performed first before trying TOCE procedure. However, a few patients with HCC and biliary obstruction have rather localized parenchymal tumour and their biliary obstruction is caused by the ingrowth of tumour from the peripheral bile duct. In these circumstances, TOCE can be tried first before biliary drainage procedure with great caution and careful post-procedure monitoring of patients.
Hepatic arterial embolization or chemoembolization has been reported to induce hepatic abscess or septicaemia in humans. These septic complications may result in death. In our series, mainly composed of HCCs and disease treated without the administration of prophylactic antibiotics, 10 (1.1%) of 942 procedures were associated with septicaemia. Fatal sepsis was documented in only three patients with predisposing factors. Therefore, in HCC, routine administration of prophylactic antibiotics is not advocated because of the low incidence of septic complications and their exclusive occurrence in patients with predisposing factors. Necrosis of the tumour and normal liver parenchyma after TOCE can be a bed for secondary infection. If extensive tumour and normal parenchymal necrosis is anticipated after TOCE, antibiotics can be used prophylactically. Recently, we have conducted a retrospective study to find out predisposing factors to liver abscess based on 6,202 TOCE procedures in 2,190 patients with hepatic tumours.[17] Liver abscess developed in 14 patients (0.6%): in 3 of 263 (1.1%) patients with portal vein obstruction, 3 of 54 (5.6%) patients with metastatic hepatic tumours, and 5 of 36 (13.9%) patients with biliary abnormalities. Statistical analysis demonstrated biliary abnormalities and metastatic tumours as significant predisposing factors for liver abscess (p < 0.05). Among all of the potential predisposing factors, bilioenteric anastomosis was the far most important one. The mortality related to liver abscess occurred in two patients (14.3%). Ten patients with liver abscess were successfully treated with parenteral antibiotics and percutaneous catheter drainage (Fig. 5). If risk factors are present, prophylactic antibiotics coverage is recommended.
Repeated performance of TOCE causes peripheral hepatic arterial occlusion due to chemical arteritis or gelatin sponge embolization, and subsequent performance of TOCE may result in intrahepatic biloma formation. Proximal bile duct necrosis rarely occurs after TOCE.
Another major category of complications of TOCE is unintentional extrahepatic embolization. When the cystic artery is occluded during hepatic artery chemoembolization, gallbladder infarction occurs in 75-90% of patients. Most cases of gallbladder infarction shows self-limited clinical course. Splenic infarction is caused by reflux of embolic materials into the splenic artery during embolization or arteriography after embolization. With coeliac artery stenosis and/or splenomegaly due to portal hypertension, the flow direction of the common hepatic artery can be reversed, and the risk of splenic artery embolization increases. Chemoembolization agents also can be directed into gastric branches via the accessory left gastric arteries that arise in the left hepatic artery and the right gastric artery that arises in the proper hepatic artery or its distal branches. If these gastric branches are not recognized before the procedure and appropriate protective measures are not taken, gastric complications are unavoidable. Supraumbilical skin rash can develop if the emulsion of iodized oil and anticancer drug flows into the falciform artery. In TOCE, at least some proportion of iodized oil injected into the hepatic artery must leave the liver through normal hepatic vasculature or an arteriovenous hepatic shunt, which causes embolization of the pulmonary vascular network in all patients who undergo TOCE. The amount of the iodized oil injected is the most important factor in the production of symptomatic pulmonary oil embolism. To prevent pulmonary oil embolism, it is recommended to use no more than 15 mL of iodized oil, administer it superselectively, and carefully observe arteriovenous hepatic shunts on initial arteriograms and during the procedure. In addition, it is emphasized that skilful angiographic technique can prevent unexpected hepatic artery injury with tip of catheters or guidewire. Proximal hepatic artery injury makes it impossible to perform complete and selective chemoembolization.
LIMITATIONS AND PROBLEMS IN LONG-TERM MANAGEMENT
Compared with the efficacy for treating encapsulated nodular HCC, TOCE has limited efficacy for treating early-stage HCC and borderline lesions and extracapsular infiltrating edge of overt HCC because they are partly supplied by the portal vein.[18] Segmental TOCE may increase therapeutic effect for those lesions.[19] In case of bulky tumours, it is impossible to administer sufficient amount of the mixture of iodized oil and anticancer drugs selectively into the feeding arteries because we can use only the limited amount of the mixture to avoid complications and the tumours has numerous feeders. Too many tumour foci including portal vein invasion and portovenous spread are also difficult to treat with TOCE.
Another important cause of treatment failure is extrahepatic collaterals supplying HCC. Development of extrahepatic collaterals supplying HCC prohibits the effective control of the tumour by hepatic artery chemoembolization. Collateral supplies to HCC can occur with or without occlusion of the hepatic arteries. In the past, hepatic artery occlusion was considered to be the major cause of extrahepatic collaterals in hepatic tumours.[20] We prospectively investigated the presence of extrahepatic collaterals in consecutive 87 patients of HCC with digital subtraction angiography. The hepatic arteries are widely patent in all patients. Extrahepatic collaterals were present in one (2%) of 61 patients with a small tumour (<-5 cm) and 12 (46%) of 26 patients with a large tumour (> 5 cm). With this data, we can conclude that extrahepatic collaterals commonly supply large HCCs without proximal hepatic artery occlusion at their initial presentation (Figs. 3, 6). The anatomic location of the tumour adjacent to the bare area and suspensory ligaments of the liver or direct invasion or adhesion to the adjacent organ seems to be the primary cause. Attenuation or occlusion of the hepatic artery may exaggerate the degree of collateral circulation. In addition, repeated chemoembolization induces extrahepatic collaterals to peripheral recurrent tumours.
These extrahepatic collaterals can be used to continue chemoembolization for the patients with hepatic artery occlusion. When both the hepatic artery and extrahepatic collaterals supply the tumour, additional chemoembolization of the extrahepatic collaterals can be tried to increase the therapeutic efficacy of hepatic artery chemoembolization[21,22] (Fig. 6). Therefore, interventional radiologists should have the knowledge about the content, causes, and suggestive imaging findings of extrahepatic collaterals to detect them early. In our institution, we analyzed a total of 280 extrahepatic collaterals demonstrated in 200 patients with HCC, which directly fed the tumour. The content and frequency of those collaterals are summarized in Table 1.
Inferior phrenic arteries 116 (41%) Omental arteries 70 (2 %) Internal mammary arteries 23( 8%) Colic branch of superior mesenteric artery 19 (7%) Adrenal arteries 19 (7%) Intercostal arteries 15 (5%) Renal capsular artery 7 (3%) Gastric arteries 6(2%) Miscellaneous 5 (2%)
Sometimes, we encountered the patients in whom HCC in the liver was well controlled by chemoembolization, but systemic metastatic lesions became the major clinical problems. Prolongation of the survival period seems to increase the chance of systemic metastasis.
Even if we do our best to perform selective chemoembolization, it is unavoidable to damage the normal liver parenchyma in a certain degree. Therefore, repeated sessions of TOCE can cause progressive deterioration of liver function of the patient. Because of tumor growth, the natural progression of liver cirrhosis and progressive deterioration of liver function, hepatic failure is the most common cause of death in patients with HCC managed by chemoembolization. Variceal bleeding due to liver cirrhosis can be the cause of death.
CONCLUSION
In small HCCs, segmental TOCE shows comparable results to surgical resection. Postoperative recurrent HCCs and HCCs in liver transplantation candidates are another good indications of TOCE. Advances in catherter technology and accumulation of knowledge about the procedure and tumor circulation may improve the effectiveness of the treatment and reduce procedure- related complications. In the early-stage of HCC, combined treatment of transcatheter and percutaneous approach is recommended to increase therapeutic effect. In the advanced stage of HCC, the treatment protocol of TOCE should be individualized according to hepatic function, the status of the portal vein, and the extent of the tumour. With this revised protocol of TOCE, the efficacy of TOCE should be re-evaluated by well-designed randomized prospective study.
REFERENCES
1. Doyon D, Mouzon A, Jourde AN, et al. L embolisation artrielle hpatique dans les tumeurs malignes du foie. Ann Radiol 1974; 17 : 593-603.
2. Yamada R, Nakatsuka H, Nakamura K, et al. Hepatic artery embolization in 120 patients with non-resectable HCC. Radiology 1983; 148 : 397-401.
3. Konno T, Maeda H, Iwai K, et al. Effect of arterial administration of high-molecular-weight anticancer agent SMANCS with lipid lymphographic agent on hepatoma. Eur J Cancer Clin Oncol 1983; 19 : 1053-65.
4. Nakamura H, Tanaka T, Hori S, et al. Transcatheter embolization of hepatocellular carcinoma: assessment of efficacy in case of resection following embolization. Radiology 1983; 147 : 401-5.
5.Mori W, Masuda M, Miyanaga T. Hepatic artery ligation and tumor necrosis in the liver. Surgery 1966; 59 : 359-63.
6. Nakamura H, Hashimoto T, Oi H, Sawada S. Transcatheter oily chemoembolization of hepatocellular carcinoma. Radiology 1989; 170 : 783-86.
7. Kan Z, Sato M, Ivancev K, et al. Distribution and effect of iodized poppyseed oil in the liver after hepatic artery embolization: experimental study in several animal species. Radiology 1993; 186 : 861-66.
8.Park JH, Chung JW, Lee SK, et al. Chemoembolization of hepatocellular carcinoma: long-term survival and prognostic factors. J Korean Radiol Society 1996; 35 : 315-323.
9.Matsui O, Kadoya M, Yoshikawa J, et al. Small hepatocellular carcinoma: Treatment with subsegmental transcatheter arterial embolization. Radiology 1993; 188 : 79-83.
10.Bronowicki J-P, Boudjem K, Chone L, et al. Comparison of resection, liver transplantation and transcatheter oily chemoembolization in the treatment of hepatocellular carcinoma. J Hepatol 1996; 24 : 293-300.
11.Lee H-S, Kim KM, Yoon JW, et al. Therapeutic efficacy of transcatheter arterial chemoembolization as compared with hepatic resection in hepatocellular carcinoma patients with compensated liver function in a hepatitis B virus-endemic area: a prospective cohort study. J Clin Oncol 2002; (in press).
12.Chung JW, Park JH, Han JK, Choi BI, Han MC. Hepatocellular carcinoma and portal vein invasion: results of treatment with transcatheter oily chemoembolization. AJR 1995; 165 : 315-321.
13.Groupe d'Etude et de Traitement du Carcinome Hepatocellulaire. A comparison of Lipiodol chemoembolization and conservative treatment for unresectable hepatocellular carcinoma. N Engl J Med 1995; 332 : 1256-61.
14.Chung JW, Park JH, Han JK, et al. Hepatic tumors: predisposing factors for complications of transcatheter oily chemoembolization. Radiology 1996; 198 : 33-40.
15.Takayasu K, Shima Y, Muramatsu H, et al. Hepatocellular carcinoma: treatment with intraarterial iodized oil with or without chemotherapeutic agents. Radiology 1987; 162 : 345-351.
16.Yamashita Y, Torashima M, Oguni T, et al. Liver parenchymal changes after transcatheter arterial embolization therapy for hepatoma: CT evaluation. Abdom Imaging 1993; 18 : 352-356.
17.Song SY, Chung JW, Han JK, et al. Liver abscess after transcatheter oily chemoembolization for hepatic tumors: incidence, predisposing factors, and clinical outcome. J Vasc Intervent Radiol 2001; 12 : 313-320.
18.Takayasu K, Wakao F, Moriyama N, et al. Response of early-stage hepatocellular carcinoma and borderline lesions to therapeutic arterial embolization. AJR 1993; 160 : 301-306.
19.Matsuo N, Uchida H, Nishimine K, et al. Segmental transcatheter hepatic artery chemoembolization with iodized oil for hepatocellular carcinoma : antitumor effect and influence on normal tissue. 1993; 4 : 543-549.
20.Charnsangavej C, Chuang VP, Wallace S, Soo C, Bowers T. Angiographic classification of hepatic collaterals. Radiology 1982; 144 : 485-494.
21.Kim JH, Chung JW, Park JH, Choi BI, Han MC. Transcatheter arterial embolization of the internal mammary artery in hepatocellular carcinoma. JVIR 1995; 6 : 71-77.
22.Chung JW, Park JH, Han JK, Choi BI, Kim TK, Han MC. Transcatheter oily chemoembolization of the inferior phrenic artery in hepatocellular carcinoma : the safety and potential therapeutic role. JVIR 1998; 9 : 495-500.
 |
