| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Nagasaki University School of Medicine, 1-7-1
Sakamoto, Nagasaki 852-8501, Japan.
2 Department of Radiology, Inoue Hospital, 8-9 Takaramachi, Nagasaki 852-0045,
Japan.
3 Department of Radiology, National Nagasaki Medical Center, 2-1001-1 Kubara,
Omura 856-0835, Japan.
4 Department of Radiology, Nagasaki Municipal Hospital, 6-39 Shinchi-machi,
Nagasaki 850-0842, Japan.
5 Second Department of Surgery, Nagasaki University School of Medicine, Nagasaki
852-8501, Japan.
6 Second Department of Internal Medicine, Nagasaki University School of
Medicine, Nagasaki 852-8501, Japan.
7 Department of Radiology, Yamaguchi University School of Medicine, 1-1-1
Kogushi, Ube, Yamaguchi 755-8505, Japan.
Received June 17, 2002;
accepted after revision January 17, 2003.
Address correspondence to I. Sakamoto.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Nine hundred seventy-two patients with hepatocellular carcinoma (n = 920) or metastatic liver tumor (n = 52) underwent chemoembolization during a 12-year period beginning in January 1989. We retrospectively reviewed the medical records and follow-up radiographs of chemoembolization and analyzed the risk factors associated with the development of intrahepatic biloma.
RESULTS. Intrahepatic biloma developed after chemoembolization in 35 patients (3.6%, 35/972) in our series. The incidence of intrahepatic biloma formation in patients with metastatic liver tumor (9.6%, 5/52) was higher than that in patients with hepatocellular carcinoma (3.3%, 30/920) (p < 0.05, Fisher's exact test). The incidence of intrahepatic biloma formation in patients with hepatocellular carcinoma was statistically higher in patients with main tumor size of less than 5 cm and in those with the presence of intrahepatic bile duct dilatation. Technique-related risk factors such as injection site of drugs, repeated chemoembolization with frequency of less than 3 months, and regimen of chemoembolization significantly influenced the incidence of biloma formation in patients with hepatocellular carcinoma. No patient died of infected biloma or septicemia, but one patient died of hepatic failure 2 months after chemoembolization.
CONCLUSION. Biloma formation was significantly more prevalent in the metastatic lesion group than in the hepatocellular carcinoma group. Significant prognostic factors for biloma formation in patients with hepatocellular carcinoma were tumor size of less than 5 cm, bile duct dilatation, proximal injection site, repeated injection with frequency of less than 3 months, and injection of a suspension of anticancer drugs.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Informed consent was obtained from all patients for each procedure. Chemoembolization was usually performed as follows: After insertion of a 4- to 5-French introducer sheath via the right or left femoral artery, a 4- to 5-French catheter was inserted into the hepatic artery. After evaluation of the feeding arteries on hepatic angiography, chemoembolization was subsequently performed with or without using a 3-French microcatheter. Depending on the tumor size and the number of tumor nodules and feeding arteries, the tip of the catheter or microcatheter was advanced into the proper, right, left, segmental, or subsegmental hepatic artery. Chemoembolization was initially performed by injection of a mixture of anticancer drugs and iodized oil (2–17 mL) or a suspension of the two, followed by injection of gelatin sponge particles. One or two of the following anticancer drugs were used for each procedure: doxorubicin hydrochloride (Adriacin, 10–30 mg, Kyowa Hakko Kogyo, Tokyo, Japan), epirubicin hydrochloride (Farmorubicin, 10–30 mg, Kyowa Hakko Kogyo, Tokyo, Japan), mitomycin C (Mitomycin C, 10–20 mg, Kyowa Hakko Kogyo, Tokyo, Japan), and cisplatin (Randa, 25–100 mg, Nippon Kayaku, Tokyo, Japan). A mixture of anticancer drugs and iodized oil was prepared as follows: anticancer drugs were dissolved in 5 mL of a fluid that was prepared from a water-soluble contrast medium and distilled water. Dissolved anticancer drugs and iodized oil were then drawn separately into syringes interconnected with a three-way stopcock and emulsified by pushing each syringe in alternation. A suspension, which was composed of anticancer drugs and iodized oil with phosphatidylcholine as a dispersing stabilizer, was prepared before chemoembolization to increase the intratumoral concentration of the anticancer agent and to decrease side effects by the gradual release of anticancer drugs from the suspension. The suspension of anticancer drugs was prepared in the following manner. In Japan, powder of doxorubicin hydrochloride, epirubicin hydrochloride, and mitomycin C are commercially available, whereas cisplatin powder is not commercially available. Therefore, we initially prepared the powder of cisplatin from the commercially available cisplatin solution. Then, phosphatidylcholine and iodized oil were mixed in an agate mortar, and the mixture was collected in a vial with the powder of anticancer drugs. The mixture of phosphatidylcholine, iodized oil, and the powder of anticancer drugs in a vial was finally dispersed in an ultrasonic cleaner for 30 min.
Seven hundred fifty of 920 patients with hepatocellular carcinoma were treated with gelatin sponge and a mixture of anticancer drugs and iodized oil, whereas the remaining 170 patients were treated with a gelatin sponge and a suspension of them. All the patients with metastatic liver tumor were treated with gelatin sponge and a mixture of anticancer drugs and iodized oil. Prophylactic antibiotics were administered in all patients for 3–5 days after the procedure.
Follow-up CT was routinely performed 2 weeks and 1, 2, 3, and 6 months after chemoembolization and at 6-month intervals thereafter. Intrahepatic biloma was diagnosed when at least one of the following signs was seen on follow-up CT: round, solitary, or multiple cystic lesions with or without segmental bile duct dilatation; a branching appearance of hypoattenuating area along the Glisson's sheath simulating dilatation of the intrahepatic bile duct; or a subcapsular fluid collection of hypoattenuating density similar to the bile duct. We retrospectively reviewed the medical records and follow-up radiographs of 2540 chemoembolization sessions and analyzed the relationship between the incidence of intrahepatic biloma formation and various risk factors such as the nature of the tumor (primary vs secondary), Child-Pugh class, main tumor size, presence of portal vein tumor thrombus, injection site of embolic materials, total number of chemoembolization sessions, time interval between each chemoembolization session in patients with more than one procedure, and a regimen of chemoembolization. Additionally, the radiographic and pathologic findings of the intrahepatic bilomas and the clinical course of the patients with intrahepatic biloma were also evaluated. The chi-square test or Fisher's exact test was used to evaluate the relationship between the incidence of intrahepatic biloma formation and various risk factors. A p value of less than 0.05 was considered to be statistically significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
In patients with hepatocellular carcinoma, potential risk factors such as a tumor smaller than 5 cm (p < 0.05, chi-square test) and dilatation of intrahepatic bile duct (p < 0.05, Fisher's exact test) influenced the incidence of intrahepatic biloma formation. However, other potential risk factors such as Child-Pugh classification and presence of portal vein tumor thrombus did not influence the incidence of intrahepatic biloma formation.
In patients with hepatocellular carcinoma, technical-related risk factors such as proximal injection of drugs (p < 0.01, chi-square test) and repeated injection with a frequency of less than 3 months (p < 0.01, chi-square test) significantly influenced the incidence of intrahepatic biloma formation. However, the frequency of chemoembolization did not influence the incidence of intrahepatic biloma formation. The relationship between predisposing risk factors and intrahepatic biloma formation in patients with hepatocellular carcinoma is summarized in Table 1.
|
In patients with metastatic liver tumors, the incidence of intrahepatic biloma formation was higher in patients with tumors of less than 5 cm (3/20, 15%) than in those with tumors of 5 cm or larger (2/32, 6.3%) and higher in patients with dilatation of the intrahepatic bile duct seen on preprocedural CT (4/43, 9.3%) than in patients without dilatation of intrahepatic bile duct (26/877, 3.0%). However, those differences were not statistically significant. In patients with metastatic liver tumor, technique-related risk factors such as injection site of drugs, frequency of chemoembolization, and repeated injection with frequency of less than 3 months did not influence the incidence of intrahepatic biloma formation. The relationship between predisposing risk factors and intrahepatic biloma formation in patients with metastatic liver tumor is summarized in Table 2.
|
Intrahepatic biloma developed in 22 (3.7%) of 590 patients receiving less than 5 mL of iodized oil, in 10 (3.4%) of 290 patients receiving at least 5 mL but less than 10 mL of iodized oil, in three (3.8%) of 80 patients receiving at least 10 mL but less than 15 mL of iodized oil, and in none of 12 patients receiving 15 mL or more of iodized oil. Therefore, the amount of injected iodized oil did not influence the incidence of intrahepatic biloma formation (Table 3). The incidence of intrahepatic biloma formation in patients with hepatocellular carcinoma treated with a suspension (14/170, 8.2%) (Figs. 1A, 1B, 1C, 1D, 1E, 1F and 2A, 2B) was statistically higher (p < 0.01, chi-square test test) than those treated with a mixture (16/750, 2.1%). The relationship between the regimen of chemoembolization and intrahepatic biloma formation in patients with hepatocellular carcinoma or metastatic liver tumor is summarized in Table 4.
|
|
In at least seven of the 35 patients with intrahepatic biloma, the catheter had been wedged into the hepatic artery because of small arterial size or severe arterial spasm when the chemoembolic agent was infused (Figs. 1A, 1B, 1C, 1D, 1E, 1F).
In 30 patients, intrahepatic biloma developed within 2 months of chemoembolization. In the remaining five patients, intrahepatic biloma developed 3 or 4 months after chemoembolization. In the seven patients in whom intrahepatic biloma developed after segmental or subsegmental chemoembolization, biloma was located in the same segment as the hepatocellular carcinoma. In 28 patients in whom intrahepatic biloma developed after proximal chemoembolization, biloma was located in various segments of the liver: the posterior segment in the right lobe (n = 20) and segment II in the left lobe (n = 7) were the most frequent locations (Table 5).
|
On CT, intrahepatic biloma appeared as a round, solitary, or multiple cystic area in 24 patients; as a branching appearance of hypoattenuating area along the Glisson's sheath simulating dilatation of the intrahepatic bile duct in eight patients; or as a subcapsular fluid collection with intrahepatic bile duct dilatation in three patients. Endoscopic retrograde cholangiography or percutaneous transhepatic cholangiography performed in five patients showed stenosis of the proximal portion of the hepatic duct.
In 21 patients who were treated conservatively because of no clinical symptoms, biloma remained unchanged in 11, disappeared in seven (Figs. 2A, 2B), and shrank in the remaining three during the follow-up period. In nine patients who were treated with percutaneous drainage because of fever and other signs of infection, biloma disappeared in five and shrank in the remaining four. One of the nine patients died of hepatic failure 2 months after chemoembolization, whereas the clinical course was uneventful in the remaining eight patients. Five patients underwent hepatectomy because of recurrent hepatocellular carcinoma or signs of infection. The postoperative course was uneventful in all these patients. Treatment and outcome of intrahepatic biloma are summarized in Table 6.
|
Pathologic specimens were obtained in six patients. Pathologic findings included intrahepatic bile duct necrosis, bile leakage through the Glisson's sheath, coagulation necrosis of the liver parenchyma adjacent to bile leakage, and thrombosis and coagulation necrosis of small arterial branches adjacent to the necrotic bile duct (Figs. 1A, 1B, 1C, 1D, 1E, 1F and 3A, 3B, 3C).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In contrast to the normal liver parenchyma, the intrahepatic bile ducts do not have a dual blood supply and are fed exclusively from the hepatic arterial branches that give off a vascular plexus (peribiliary capillary plexus) around the bile ducts. Therefore, ischemia of the intrahepatic bile ducts can easily occur after chemoembolization [17, 18]. The possible mechanism of biloma formation after chemoembolization is considered to be the development of peripheral bile duct necrosis with bile leakage caused by microvascular damage of the peribiliary capillary plexus [6, 8, 18]. This mechanism was supported by the following pathologic findings in our study: intrahepatic bile duct necrosis, bile leakage through the Glisson's sheath, coagulation necrosis of the liver parenchyma adjacent to bile leakage, and thrombosis and coagulation necrosis of the small arterial branches adjacent to the necrotic bile duct.
The blood supply to the bile duct system was evaluated by Northover and Terblanche in 1979 [19]. They found that the right and left hepatic ducts were supplied by numerous small vessels from the right and left hepatic arteries. The supraduodenal duct, defined as the common hepatic duct and upper common bile duct, is supplied by axial vessels that, in 60% of cases, run upward from the retroduodenal or gastroduodenal arteries, but in 38% of cases these vessels arise from the right hepatic artery. Hence, necrosis of the right and left hepatic ducts, and rarely the supraduodenal duct, can occur after chemoembolization when a catheter tip is placed at the proximal portion of the right or left hepatic artery. In our series, stenosis of the extrahepatic bile duct was shown with endoscopic retrograde cholangiography or percutaneous transhepatic cholangiography in five patients. In all five patients, the chemoembolic agents were injected into the proximal portion of the right, left, or both hepatic arteries. Stenosis of the extrahepatic bile duct could be another contributing factor for the development of intrahepatic biloma because stenosis may induce intrahepatic bile stasis and cholangitis. Therefore, injection of the chemoembolic agents into the proximal portion of the hepatic artery should be avoided except in cases in which distal catheterization attempts are technically unsuccessful, even with the use of a microcatheter.
The incidence of intrahepatic biloma formation was higher in patients with metastatic liver tumor than in those with hepatocellular carcinoma. In hepatocellular carcinoma, high concentrations of chemoembolic agents in the tumor tissue after infusion can be achieved because the arteries that feed the tumor and intratumoral blood space are mostly dilated. On the other hand, the feeding artery and the intratumoral blood space of metastatic liver tumor are not usually dilated, thus decreasing the intratumoral concentration of chemoembolic agents. This may result in higher concentration of the chemoembolic agents in the surrounding liver parenchyma and initiation of biliary epithelial damage. Therefore, the difference in tumor vascularity between hepatocellular carcinoma and metastatic liver tumor may be the cause of the difference in the incidence of intrahepatic biloma formation between the two groups. Additionally, in Japan, hepatocellular carcinoma is mostly seen in cirrhotic livers that are known to have dilatation of the perivascular plexus, which can act as a portoarterial shunt and compensate for the decreased arterial flow [20]. This can be another factor for lower incidence of intrahepatic biloma in hepatocellular carcinoma.
In a large liver tumor, the intratumoral blood space of the tumor is usually dilated, thus increasing the intratumoral concentration of chemoembolic agents. This may result in lower accumulation of the chemoembolic agents in the nontumoral liver parenchyma with a low chance of biliary damage. In our series, the incidence of intrahepatic biloma was significantly lower in patients with hepatocellular carcinomas of 5 cm or larger.
In patients with liver tumors, intrahepatic bile ducts are often dilated as a result of various causes such as tumor compression, bile duct stones, and surgical intervention. Bile duct dilatation can induce intrahepatic bile stasis and cholangitis with increased likelihood of development of intrahepatic biloma. In our series, patients with hepatocellular carcinoma had a significantly higher incidence of intrahepatic biloma when intrahepatic bile duct dilatation was seen on preprocedural CT.
A suspension of anticancer drugs and iodized oil has been recently used in transcatheter arterial infusion or embolization. Compared with a mixture, the suspension has an advantage of gradual release of anticancer drugs achieved by the use of phosphatidylcholine as a dispersing stabilizer. This can result in a high concentration of anticancer drugs in the tumor tissue for a long period of time [21, 22]. Moreover, the use of the suspension can reduce side effects by keeping the concentration of anticancer drugs in the blood low. However, this may also expose the nontumoral liver parenchyma to cytotoxic agents for a long period of time and may cause biliary epithelial damage. In our series, the incidence of intrahepatic biloma after chemoembolization was higher in patients with hepatocellular carcinoma treated with a suspension of anticancer drugs and iodized oil than in those treated with a mixture.
According to previous reports, intrahepatic biloma is often seen when chemoembolization is carried out repeatedly over a short period of time [8, 23]. This is understandable considering the fact that repetition of chemoembolization causes occlusion of the peripheral hepatic artery due to chemical vasculitis or gelatin sponge embolization, resulting in reduction of the vascular bed in nontumoral liver parenchyma. In our series, frequency of chemoembolization did not influence the incidence of intrahepatic biloma formation. However, in patients in the hepatocellular carcinoma group who underwent more than one procedure, the incidence of intrahepatic biloma formation was significantly higher in those who underwent repeated chemoembolization with frequency of less than 3 months than in those with frequency of 3 months or more. Accordingly, we recommend a time interval of more than 3 months in repeated chemoembolization.
The amount of the iodized oil is considered to be one of the most important factors in the development of serious complications after chemoembolization. However, a safe dose of iodized oil to nontumoral liver parenchyma has not been determined as yet because various factors, such as the size and vascularity of the tumor, the severity of liver dysfunction, the range of embolization, or the degree of portal invasion, affect the optimal dose of the iodized oil. In our study, amounts of iodized oil injected during chemoembolization did not influence the incidence of biloma formation. According to Nakamura et al. [24], one guideline is whether the portal vein can be retrogradely opacified with intraarterially infused iodized oil that can pass through the arteriosinusoidal or arterioportal shunt into the portal branches. The authors emphasized that infusion of embolic materials should be stopped to avoid infusion of an excessive dose of chemoembolic agent if the portal vein is visible on fluoroscopy. In our series, 35 patients with intrahepatic biloma had undergone chemoembolization with a mixture or suspension that was composed of anticancer drugs and 1–15 mL of iodized oil. In 22 of the 35 patients, the doses of iodized oil that induced intrahepatic biloma were only less than 5 mL. Therefore, even these small doses of iodized oil can cause intrahepatic biloma, especially when gelatin sponges enhance the effect of iodized oil.
In at least seven of the 35 patients with intrahepatic biloma, a catheter had been wedged into the hepatic artery as a result of the small arterial size or severe arterial spasm during chemoembolization. The relationship between catheter wedging during chemoembolization and the incidence of intrahepatic biloma formation cannot be exactly assessed because of the retrospective nature of our study. However, catheter wedging may have resulted in infusion of an excessive dose of chemoembolic agent in nontumoral liver parenchyma. The use of a microcatheter is recommended to prevent wedging of the catheter into the hepatic artery [12]. Additionally, to reduce spasm of the hepatic artery, a small amount of 2% lidocaine or vasodilator such as nicardipine hydrochloride and nitroglycerin can be intraarterially injected before a catheter is advanced into the hepatic artery.
Most intrahepatic bilomas develop within 2 months of chemoembolization. However, some may also develop at a later stage. Therefore, close follow-up of imaging is necessary, especially when alkaline phosphatase or bilirubin levels are elevated and clinical signs suggest infection.
Intrahepatic biloma had a tendency to occur in the posterior segment and segment II, which are the most gravitationally dependent portions of the each lobe of the liver with the patient in the supine position. This suggests that the distribution of chemoembolic agents in the bloodstream depends not only on the size and vascularity of liver tumors but also on the gravitational gradient.
Twenty-one of the 35 patients with biloma were treated conservatively because neither severe signs of infection nor interval increase of biloma was seen. In all 21 patients, intrahepatic biloma remained unchanged, shrank, or disappeared, and no symptoms or clinical signs of sepsis were seen during the follow-up period. In the remaining 14 patients, percutaneous drainage or segmental hepatectomy combined with the use of antibiotics was performed because of moderate to severe signs of infection or interval increase of biloma. In all but one of these patients, long-term palliation of infected biloma and cholangitis was achieved by these treatments. One patient died of hepatic failure 2 months after chemoembolization, although signs of infection subsided after percutaneous drainage. Therefore, in our series, no patient died of infected biloma or septicemia. According to our results, the outcome of intrahepatic biloma is not poor if prompt percutaneous drainage or surgical treatment is performed in addition to the use of antibiotics in patients in whom the biloma shows signs of infection or interval increase in size.
In conclusion, biloma formation was significantly more prevalent in the metastatic lesion group than in the hepatocellular carcinoma group. Significant prognostic factors for biloma formation in patients with hepatocellular carcinoma were tumor size smaller than 5 cm, bile duct dilatation, proximal injection site, repeated injection with frequency of less than 3 months, and injection of a suspension of anticancer drugs. Intrahepatic biloma can be conservatively treated in most cases. However, percutaneous drainage or surgery should be promptly performed in patients with more serious conditions such as progression of infected biloma or sepsis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. H. Kim, H. K. Lim, D. Choi, W. J. Lee, S. H. Kim, M. J. Kim, S. J. Lee, and J. H. Lim Changes in Bile Ducts after Radiofrequency Ablation of Hepatocellular Carcinoma: Frequency and Clinical Significance Am. J. Roentgenol., December 1, 2004; 183(6): 1611 - 1617. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
