Preoperative Percutaneous Transhepatic Portal Vein Embolization With Ethanol Injection

Transcription

1 Vascular and Interventional Radiology Original Research Sakuhara et al. Embolization of Vein With Ethanol Injection Vascular and Interventional Radiology Original Research Yusuke Sakuhara 1 Daisuke Abo 1 Yu Hasegawa 1 Tadashi Shimizu 2 Toshiya Kamiyama 3 Satoshi Hirano 4 Daisuke Fukumori 5 Takeshi Kawamura 4 Yoichi M. Ito 6 Khin Khin Tha 1 Hiroki Shirato 1 Satoshi Terae 1 Sakuhara Y, Abo D, Hasegawa Y, et al. Keywords: ethanol, hepatic resection, hypertrophy, liver volume, portal vein embolization DOI:1.2214/AJR Received January 16, 211; accepted after revision August 3, Department of Radiology, Hokkaido University Graduate School of Medicine, N-15, W-7, Kita-ku, Sapporo, Hokkaido 6-838, Japan. Address correspondence to Y. Sakuhara (yusaku@med.hokudai.ac.jp). 2 Faculty of Health Sciences, Hokkaido University Graduate School of Health Sciences, Hokkaido, Japan. 3 Department of General Surgery, Hokkaido University Graduate School of Medicine, Hokkaido, Japan. 4 Department of Surgical Oncology, Hokkaido University Graduate School of Medicine, Hokkaido, Japan. 5 Department of Gastroenterological Surgery Medicine, Fukuoka University Hospital, Fukuoka, Japan. 6 Department of Clinical Trial Management, Hokkaido Organization for Translational Research, Hokkaido University Graduate School of Medicine, Hokkaido, Japan. AJR 212; 198: X/12/ American Roentgen Ray Society Preoperative Percutaneous Vein Embolization With Ethanol Injection OBJECTIVE. The purpose of this article is to evaluate the feasibility and efficacy of preoperative with ethanol injection. MATERIALS AND METHODS. We retrospectively evaluated 143 patients who underwent. Hypertrophy of the future liver remnant was assessed by comparing the volumetric data obtained from CT image data before and after. The evaluation of effectiveness was based on changes in the absolute volume of the future liver remnant and the ratio of the future liver remnant to the total estimated liver volume. RESULTS. Ten of 143 patients (7.%) underwent additional embolization because of recanalization and insufficient hypertrophy of the future liver remnant. The mean increase in the ratio of the future liver remnant was 33.6% (p <.1), and the mean ratio of future liver remnant to total estimated liver volume increased from 34.9% to 45.7% (p <.1). Although most of the patients complained of pain after ethanol injection, they were gradually relieved of pain in a few minutes by conservative treatment. Fever (38 39 C) was reported after 47 of 151 (31.1%) sessions and was resolved within a few days. Transient elevation of the liver transaminases was observed after the procedures and resolved within about a week. Major complications occurred in nine of 151 (6%) percutaneous transhepatic portal vein embolization sessions, but no patients developed hepatic insufficiency or severe complications precluding successful resection. One hundred twenty patients underwent hepatic resection, and two patients developed hepatic failure after surgery. CONCLUSION. Preoperative with ethanol is a feasible and effective procedure to obtain hypertrophy of the future liver remnant for preventing hepatic failure after hepatectomy. O ne of the major reasons for unresectability of the liver is that the remnant liver volume is insufficient to support postoperative liver function. Liver insufficiency after hepatectomy is one of the most serious complications in patients undergoing major hepatic resection for hepatobiliary diseases [1 9]. It has been shown that liver failure is directly related to the functional volume of the liver remnant [1, 2]. Transcatheter portal vein embolization is performed widely for inducing atrophy of the lobe to be resected and compensatory hypertrophy of the contralateral lobe [1 14]. This procedure is considered to reduce the risks of patients who are potential candidates for major hepatic resection. In previous reports, various embolization materials have been used in portal vein embolization [1 1]. Recently, N-butyl cyano- acrylate (NBCA) with iodized oil or spherical microparticles has been widely used for portal vein embolization because of its long-term embolic effect. The effectiveness of embolization by NBCA leads to fast and reliable hypertrophy; however, it frequently induces an inflammatory process that may increase the difficulty of the subsequent surgical procedure, and it is also difficult to use in clinical practice [7, 15, 16]. Spherical microparticles also seem to be as effective as NBCA, and are easy to use and cause little periportal reaction [16]; however, they are not available in Japan because they have not been approved by the Ministry of Health. Ethanol not only causes thrombus formation but also has the advantage of strong contact destructivity. It is diluted immediately, and the systemic cytotoxicity is negligible [17]. Shimamura et al. [3] reported the results of 914 AJR:198, April 212

2 Embolization of Vein With Ethanol Injection seven patients who underwent portal vein embolization with ethanol. However, results from studies involving a large series of patients are lacking, and the feasibility and efficacy of portal vein embolization with ethanol remain uncertain. To our knowledge, this is the first report on the feasibility and efficacy of ethanol in the change in the liver volume and function and complications after. Although Nagino et al. [18] have reported the efficacy of percutaneous transhepatic portal vein embolization, their results are the combined data of cases using ethanol and cases using fibrin glue. There are two methods for portal vein embolization according to the access route into the hepatic portal vein. One is the percutaneous transhepatic approach (percutaneous transhepatic portal embolization), and the other is the transileocolic approach (transileocolic portal embolization) [1 14]. At our institution, percutaneous transhepatic portal vein embolization is the preferred method of portal vein embolization because percutaneous transhepatic portal vein embolization is less invasive and can be performed under local anesthesia. The purpose of this study was to evaluate the feasibility and efficacy of preoperative with ethanol injection. Materials And Methods Patients The study protocol was approved by the institutional review board, and informed consent was waived. A retrospective review of data for patients who underwent percutaneous transhepatic portal vein embolization was performed. We reviewed the cases of 143 consecutive patients scheduled to undergo hepatic resection for hepatobiliary disease between October 1999 and April 29. The patients liver volumes were semiautomatically measured using contrast-enhanced CT image data (volume data or 5-mm-thickness axial image data) supervised by a surgeon, and they underwent preoperative percutaneous transhepatic portal vein embolization to induce selective hepatic hypertrophy of the remaining hepatic lobe. The volume limit for hepatectomy has not been established and may vary among patients or institutions [1 3, 8, 19 21]. At our institution, in general, the indication for percutaneous transhepatic portal vein embolization was that the future liver remnant volume of the patient was less than 4 cm 3 and the ratio of future liver remnant to total estimated liver volume was less than 4% in patients whose indocyanine green (ICG) retention rate at 15 minutes before portal vein embolization was less than 15%. These criteria were not sufficient if the liver function parameters (e.g., ICG retention rate at 15 minutes, 99m Tc diethylenetriamine pentaacetic acid galactosyl human serum albumin scintigraphy, and chronic liver disease) showed the potential for liver failure after hepatectomy. This study included 31 patients whose future liver remnant volume was larger than 4 cm 3 or whose ratio of future liver remnant to total estimated liver volume was larger than 4%, but for whom the surgeon requested portal vein embolization to increase the future liver remnant volume and ratio of future liver remnant to total estimated liver volume as much as possible and improve the safety of liver resection. The patients were 96 men and 47 women (mean [± SD] age, 65.5 ± 8.5 years; range, years). All patients except four had malignant diseases: hilar cholangiocarcinoma in 68 patients, hepatocellular carcinoma in 27 patients, gallbladder carcinoma in 23 patients, intrahepatic cholangiocarcinoma in 16 patients, metastases in four patients, and carcinoid in one patient. Three of the remaining four patients had echinococcosis, and one had cavernous hemangioma. Twenty-eight patients had underlying chronic liver disease (Table 1). Five of the 27 patients with hepatocellular carcinoma underwent transcatheter arterial chemoembolization, and all four patients with metastatic carcinoma underwent systemic chemotherapy before percutaneous transhepatic portal vein embolization. No other patients had previously undergone systemic chemotherapy or transcatheter therapy. The patient characteristics are summarized in Table 1. Exclusion criteria were as follows: unresectable extrahepatic metastatic disease, lymphadenopathy indicative of unresectable metastatic disease, presence of portal vein occlusion, uncorrectable coagulopathy, severe ascites, and alcohol intolerance as determined by history of signs and symptoms, such as severe flushing, tachycardia, headaches, and hypertension following alcohol intake. Except for true coagulopathy, no criteria for hepatic enzyme levels were used to exclude patients from undergoing percutaneous transhepatic portal vein embolization or surgical resection. However, if the patients had hyperbilirubinemia secondary to biliary obstruction, the biliary system of the remnant hepatic segments was endoscopically or percutaneously decompressed before percutaneous transhepatic portal vein embolization, because preoperative biliary drainage reduces the risks associated with major hepatectomy [22]. Helical CT scans (including triple phase dynamic imaging) of the abdomen and pelvis were performed to diagnose the extent of disease, the presence of distant metastatic disease, and the presence of portal venous or hepatic arterial disorder. interpretation of the CT images, the CT scans were evaluated by at least one interventional radiologist and the referring surgeon. With CT image data, future liver remnant and total estimated liver volume were calculated before and at approximately 2 4 weeks after to determine the degree of hypertrophy. Volumetry before percutaneous transhepatic portal vein embolization was performed after transcatheter arterial chemoembolization or systemic chemotherapy. The volume data used for this study were obtained from the medical records. Evaluation of the preoperative future liver remnant and total estimated liver volume was performed in the surgical faculty by the referring surgeon. The volume of the intrahepatic tumor was measured and subtracted from the total volume of the liver. The percentage increase in the ratio of the future liver remnant (FLR) after percutaneous transhepatic portal vein embolization (PTPE) was calculated according to the following formula: volume of the FLR after PTPE-volume of the FLR before PTPE volume of the FLR before PTPE 1% (1) Percutaneous Vein Embolization Procedure Before percutaneous transhepatic portal vein embolization, 15 mg of pentazocine and.5 mg of atropine sulfate were administrated intramuscularly. Prophylactic antibiotics were not administered except for management of cholangitis. The ipsilateral approach was routinely used [12, 23, 24] (Figs. 1 and 2), with the contralateral approach used only for patients for whom the ipsilateral approach was judged to be unsuitable by interventional radiologists. With sonographic guidance, the intrahepatic portal vein was punctured with an 18-gauge needle (Needle for Ultrasonically Guided Puncture, Create Medic) under local anesthesia. A guidewire was inserted into the portal vein through the needle, followed by the introduction of a 5.5-French sheath introducer (Introducer Set, Medikit). direct portography was performed to evaluate the anatomy and pressure of the portal system, we performed selective portography with a balloon occlusion catheter (Selecon MP Catheter II, Terumo). The method of ethanol injection was as follows. First, a test injection of the contrast material with balloon occlusion was performed at a rate of.5 1. ml/s to confirm the absence of reflux. Balloon occlusion was performed, and contrast material was injected until the targeted portal branches AJR:198, April

3 Sakuhara et al. TABLE 1: Demographic Characteristics of 143 Patients Characteristic Value Age (y), mean ± SD (range) 65.5 ± 8.5 (42 84) Sex Male 96 (67.1) Female 47 (32.9) Type of disease Hilar cholangiocarcinoma 68 (47.6) Hepatocellular carcinoma 27 (18.9) Gallbladder carcinoma 23 (16.1) Intrahepatic cholangiocarcinoma 16 (11.2) Metastases 4 (2.8) Alveolar echinococcosis 3 (2.1) Carcinoid tumor 1 (.7) Cavernous hemangioma 1 (.7) Coexisting hepatic disease None 115 (8.4) Hepatitis B virus infection 19 (13.3) Hepatitis C virus infection 8 (5.6) Autoimmune hepatitis 1 (.7) Note Except where noted otherwise, data are no. (%) of patients. were enhanced. The balloon was then deflated, and an equal amount (equivalent to the previously injected contrast material) of.5% lidocaine was injected. Finally, balloon occlusion was repeated, and an equal amount of ethanol (equivalent to the previously injected contrast material or.5% lidocaine) was injected. The balloon was deflated after 5 minutes, and whether the targeted vessels were completely embolized was determined by test portography through a manual injection of contrast medium. If embolization was not complete, the ethanol injections were repeated in the same way. repeating embolization until hepatic parenchymal enhancement had disappeared, direct portography was performed to confirm the final results of. When the patient reported severe pain, 15 mg of pentazocine, 5 mg of flurbiprofen axetil, 1 2 mg of butorphanol tartrate, or 5 1 μg of fentanyl was administrated IV. To terminate the procedure, the 5.5-French sheath was extracted by packing the puncture tract with gelatin sponge torpedoes (Spongel, Astellas Pharma) in 141 patients or metallic coils in two patients. Packing was needed to prevent hemorrhage from the punctured portal vein because the tract along which the sheath introducer and catheter were introduced remained patent after their removal and the blood leaked through it. Injury to peripheral arteries also induces hemorrhage, which needs packing. packing, the leakage of blood ceased. Patient Follow-Up Examination Evaluation of postembolization syndrome and biochemical changes before and after percutaneous transhepatic portal vein embolization included a review of patient symptoms and clinical signs and laboratory data, including WBC count, platelet count, and serum levels of total bilirubin, aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, and γ-glutamyl transpeptidase. Complications after percutaneous transhepatic portal vein embolization and surgery were recorded according to information obtained from the referring physician and medical records. Contrast-enhanced CT images were obtained at 1 and 2 weeks after percutaneous transhepatic portal vein embolization to confirm portal thrombus formation, and future liver remnant and total estimated liver volume after percutaneous transhepatic portal vein embolization were calculated at approximately 2 4 weeks after percutaneous transhepatic portal vein embolization to determine the degree of hypertrophy, as mentioned previously in this article [16, 18, 22]. Surgical resection was performed by hepatobiliary surgeons. The type of surgical resection was determined after review of the patient s clinical status and radiologic study findings. If a paraaortic lymph node or distant metastasis was evident by preoperative image findings or operative speculation, hepatic resection was not performed. Postoperative monitoring was performed by the surgical staff, including daily evaluation of clinical and laboratory data. A Statistical Analysis Data are expressed as mean (± SD), unless stated otherwise. Statistical analysis was performed as indicated with a statistical analysis program package (JMP version 8., SAS). The nonparametric matched-pairs test (Wilcoxon signed rank test) was applied for bivariate data to analyze the changes in total estimated liver volume, future liver remnant, and the ratio of future liver remnant to total estimated liver volume ratio. p values less than.5 were considered to indicate significant differences. Results Percutaneous transhepatic portal vein embolization was attempted in 143 patients, and the embolized portal branches were as follows: the right branch in 132 patients; the right anterior and left branches of segments II, III, and IV in five patients; the right and left branches of segment IV in three patients; the right branch of segment VIII and the left branches of segments II, III, and IV in one patient; the right anterior branch in one patient; and the right posterior branch in one patient. Two patients underwent embolization of only one segment because the CT images and portogram proved occlusion of the right posterior or anterior branch secondary to the compression of a tumor. The ipsilateral approach was used in 139 patients, and the contralateral approach was used in four patients because of a failed ipsilateral approach in one case and to avoid tumor puncture in three cases. Additional portal vein embolization (percutaneous transhepatic portal vein embolization in eight patients and transileocolic portal embolization in two patients) was performed in 1 patients (7.%), because post- Fig. 1 Five-French triple-lumen balloon occlusion catheter. A and B, Photographs show end hole (arrow, A) at distal catheter tip and side hole (arrow, B) at proximal side of balloon. B 916 AJR:198, April 212

4 Embolization of Vein With Ethanol Injection recanalization of the targeted vessels was proven on CT or ultrasound images, or both, and their future liver remnant hypertrophy after the first percutaneous transhepatic portal vein embolization was insufficient for hepatic resection. Eight of these patients obtained sufficient hypertrophy after additional portal vein embolization. However, sufficient hypertrophy was not observed in the other two patients, despite additional portal vein embolization. The other nine patients who were also found to have recanalization did not undergo additional portal vein embolization because sufficient future liver remnant hypertrophy was obtained. The amount of injected ethanol was 12.3 ± 5.4 ml per session and 13. ± 6. ml per patient. A D Fig year-old man with hilar cholangiocarcinoma who underwent with ethanol by ipsilateral approach. A, Direct portogram followed by sheath introducer insertion via right branch of portal vein before embolization. B, Balloon occlusion catheter (white arrows) is placed into right anterior branch of portal vein, and balloon (arrowheads) is inflated. Nasobiliary catheter (black arrows) is also shown. C, Selective portogram with balloon occlusion shows right anterior branch of portal vein that was subsequently embolized with ethanol injection from distal end hole. D, Selective portogram with balloon occlusion shows right posterior branch of portal vein that was subsequently embolized with ethanol injection from proximal side hole. E, Direct portogram obtained after embolization of right portal vein shows that there is no residual flow in right branch of portal vein. Ninety-six patients who had biliary obstruction were successfully treated before with percutaneous or endoscopic drainage: endoscopic nasobiliary drainage was used in 5 patients, percutaneous transhepatic biliary drainage (PTBD) was used in 37 patients, endoscopic biliary stenting was used in three patients, PTBD combined with percutaneous transhepatic gallbladder drainage was used in three patients, endoscopic nasobiliary drainage with PTBD was used in two patients, and endoscopic biliary stenting with PTBD was used in one patient. CT volumetric data were obtained 17.3 ± 1.4 days after percutaneous transhepatic portal vein embolization in all except four patients. Two of the four patients were diagnosed B E as having unresectable extrahepatic disease (metastatic paraaortic lymph node in one patient and metastatic peritoneal nodular lesion in the other) on follow-up CT images after, and their volumetric data were not calculated. The remaining two patients, who underwent transileocolic portal embolization after the first percutaneous transhepatic portal vein embolization, were excluded from evaluation of liver volume changes because the aim of this study was to evaluate the results of percutaneous transhepatic portal vein embolization. The mean total estimated liver volume was 1217 ± 252 cm 3 before percutaneous transhepatic portal vein embolization and 1212 ± 253 cm 3 after percutaneous transhepatic portal vein embolization, and these values were C AJR:198, April

5 Sakuhara et al. Total Estimated Liver Volume (cm 3 ) Ratio of Liver Volume to Total Estimated Liver Volume (%) Mean: 1217 ± 252 cm 3 Before Percutaneous Mean: 34.9 ± 8.7 cm 3 Before Percutaneous p =.3212 p =.1 not significantly different (p =.3212) (Fig. 3). The mean future liver remnant increased from 419 ± 121 cm 3 to 542 ± 13 cm 3, and the mean absolute increase of the future liver remnant was 123 ± 91 cm 3. This increase was statistically significant (p <.1). The mean increase in the ratio of the future liver remnant was 33.6 ± 31.7%, and this increase was also statistically significant (p <.1). The mean ratio of future liver remnant to total estimated liver volume was 34.9 ± 8.7% before percutaneous transhepatic portal vein embolization and 45.7 ± 1.4% after percutaneous transhepatic portal vein embolization. The mean increase of the ratio of future liver remnant to total estimated liver volume was 1.7 ± 6.7%, and was also statistically significant (p <.1). Mean: 1212 ± 253 cm 3 Percutaneous Mean: 45.7 ± 1.4 cm 3 Percutaneous A C Future Liver Remnant Volume (cm 3 ) Mean: 419 ± 122 cm 3 Before Percutaneous p =.1 Mean: 542 ± 13 cm 3 Percutaneous Fig. 3 Laboratory values before and after percutaneous transhepatic portal vein embolization. A, Mean total estimated liver volume was 1217 ± 252 cm 3 before percutaneous transhepatic portal vein embolization and 1212 ± 253 cm 3 after percutaneous transhepatic portal vein embolization, with no significant difference between them (p =.3212). B, Mean future liver remnant increased from 419 ± 121 cm 3 to 542 ± 13 cm 3, and mean increase in ratio of future liver remnant was 33.6 ± 31.7%. Increase was statistically significant (p <.1). C, Mean ratio of future liver remnant to total estimated liver volume was 34.9 ± 8.7% before and 45.7 ± 1.4% after, with significant difference between them (p <.1). Mean increase of ratio of future liver remnant to total estimated liver volume ratio was 1.7 ± 6.7%. Although most of the patients complained of pain immediately after ethanol injection, they were gradually relieved of pain in a few minutes without any additional treatment or with administration of analgesics. Fever (38 39 C) was reported in 47 of 151 (31.1%) sessions. In three cases, fever was associated with acute obstructive cholangitis secondary to bile duct carcinoma. Antibiotics were administered in all three cases, and the initial drainage catheter was changed in one case (from endoscopic biliary stenting to endoscopic nasobiliary drainage). Fever waned within 1 3 days after conservative treatment by cooling or administration of antipyretics in the remaining patients. None of the patients had chills or bacteremia. Although transient elevation of the WBC count and serum total bilirubin and liver transaminase levels and transient decrease of the platelet count were observed after the procedure, they improved within about a week after percutaneous transhepatic portal vein embolization (Table 2 and Fig. 4). No patient developed fulminant liver insufficiency or died as a result of complications. No complications precluded successful resection. Major complications related to the procedures occurred in nine of 151 sessions (6.%) and nine of 143 patients (6.3%). Pneumothorax occurred in two patients and required tube drainage in one patient. Intrahepatic arterial hemorrhage occurred in two patients and required transcatheter arterial embolization with metallic coils in both. Subcap- B 918 AJR:198, April 212

6 Embolization of Vein With Ethanol Injection TABLE 2: Blood Test Data Before Percutaneous and Peak and Nadir Values Percutaneous Parameter Value Before Percutaneous Vein Embolization Peak Value Percutaneous Nadir Value Percutaneous WBC count (cells 1 3 /m 3 ) 6.3 ± 2.1 ( ) 8.9 ± 3. ( ) 5.3 ± 1.7 ( ) Platelet count (cells 1 3 /m 3 ) 244 ± 87 (87 572) 278 ± 96 (57 574) 177 ± 79 (1 485) Total bilirubin level (mg/dl) 1.2 ± 1. (.3 4.8) 2.1 ± 1.6 (.3 1.3).9 ±.6 (.2 3.7) Aspartate aminotransferase level (IU) 56.3 ± 54.4 (16 449) 38 ± 438 (35 312) 41.4 ± 16.4 (12 99) Alanine aminotransferase level (IU) 81. ± 81. (7.6 51) 375 ± 389 ( ) 66. ± 38.9 (6.8 24) Lactate dehydrogenase level (IU) 263 ± 23 ( ) 571 ± 534 ( ) 244 ± 198 (97 218) γ-glutamyl transpeptidase level (IU) 245 ± 247 ( ) 38 ± 249 ( ) 179 ± 157 (2 1116) Note Data are mean ± SD (range). sular hematoma of the liver occurred in two patients and required transcatheter arterial embolization in one patient. Severe thrombocytopenia occurred in one patient, who required transfusion with 2 U of platelets and then recovered. Two patients developed nontargeted portal vein thrombosis continuing from the embolized branch, which was proved on follow-up ultrasound and CT images. In both patients, portal patency and blood flow were maintained, and the thrombus was successfully removed at the time of hepatectomy after close observation on ultrasound without anticoagulation therapy. The indication for hepatic resection was decided not only by future liver remnant hypertrophy but also by liver function based on laboratory data (i.e., ICG retention rate at 15 minutes or 99m Tc diethylenetriamine pentaacetic acid galactosyl human serum albumin scintigraphy). One hundred twenty patients (83.9%) underwent planned hepatic resection: right hepatic lobectomy was performed in 88 patients, extended right lobectomy was performed in 14 patients, right trisegmentectomy was performed in 1 patients, left trisegmentectomy was performed in four patients, right hepatic lobectomy and left partial resection was performed in two patients, right anterior segmentectomy was performed in one patient, and an extended right lobectomy and left partial resection was performed in one patient. Four patients (2.8%) with extrahepatic cholangiocarcinoma underwent limited bile duct resection without hepatic resection because of insufficient liver function (n = 2) and insufficient hypertrophy of the future liver remnant (n = 2). Two patients who underwent hepatic resection died of multiorgan failure after surgery. Nineteen patients (13.3%) did not undergo any surgical resection. In 15 of these 19 patients (78.9%), unresectable malignant lesions such as paraaortic lymph node metastases (n = 6), peritoneal dissemination (n = 6), minimal hepatic metastases (n = 2), and severe perineural invasion (n = 1) were found on CT images after (2/15) or at laparotomy (13/15). In the remaining four patients (21.1%), the surgical procedure had to be canceled because of insufficient liver function (n = 2), insufficient hypertrophy of the future liver remnant (n = 1), and severe hepatic fibrosis (n = 1). Consequently, eight of 143 patients (5.5%) did not undergo the planned resection because of insufficient liver function, insufficient hypertrophy of the future liver remnant, or hepatic fibrosis. With regard to disease type, four of 16 patients (25%) with intrahepatic cholangiocarcinoma, five of 23 patients (21.7%) with gallbladder carcinoma, and six of 68 patients (8.8%) with hilar cholangiocarcinoma did not undergo surgical resection because of unresectable metastatic lesions not detected on ultrasound, CT, or MRI before percutaneous transhepatic portal vein embolization. Discussion In the current study, a high degree of absolute hypertrophy of the future liver remnant and an increase in the ratio of future liver remnant to total estimated liver volume were gained after percutaneous transhepatic portal vein embolization with ethanol. The mean increase of the future liver remnant volume was 123 ± 91 cm 3 and that for the ratio of the future liver remnant was 33.6%. The mean increase of the ratio of future liver remnant to total estimated liver volume ratio was 1.7%. In a metaanalysis involving 188 patients, Abulkhir et al. [13] reported that the mean increase in the ratio of the future liver remnant was 11.9% in the percutaneous transhepatic portal vein embolization group and 9.7% in the transileocolic portal embolization group. In other reports, absolute hypertrophy of the future liver remnant has been reported to range from 33% to 43% [8, 9, 18, 25]. Our results thus show equivalent efficacy compared with previous reports. Embolization of the vessels with ethanol injection has been shown to cause total vascular occlusion through its contact destructivity [17]. Ethanol produces sludging of blood capsules, denaturation, and coagulation of protein. When it comes into contact with the endothelium of the portal veins, it causes intimal damage, resulting in perivascular leukocyte infiltration [26]. Hepatic tissue damage rapidly occurs in the perivascular area before the development of collateral vessels after ethanol injection [17]. Injected ethanol penetrates deeply into the liver parenchyma when injected through the portal veins, and the destructive effect on the liver tissue and the liver weight increase are proportional to the dose of ethanol injected [17, 26]. Portal vein embolization is more effective than portal vein ligation at increasing the future liver remnant; the formation of collaterals between the occluded and nonoccluded portions of the liver parts seems to limit regeneration in the ligation group [27, 28]. Madoff et al. [9, 12] reported that portal vein embolization with small spherical particles provides improved hypertrophy and resection rates compared with portal vein embolization with larger nonspherical particles. They hypothesized that the smaller spherical particles flowed more distally into the portal venous system and that distal embolization limited the development of any collateral blood supply that could potentially reduce hypertrophy. We consider that the effect of peripheral vessel occlusion and the cytotox- AJR:198, April

7 Sakuhara et al. Mean Total Billrubin Level (mg/dl) Mean WBC (cells 1 3 ) Before 1 3 Days icity of ethanol prevents the development of collateral vessels and contributes to the high degree of future liver remnant hypertrophy in portal vein embolization with ethanol. Fever developed after percutaneous transhepatic portal vein embolization but improved with conservative treatment. We did not have more infectious complications, despite the fact that 96 of 143 patients (67.1%) required biliary management. It is thought to 6 9 Days Days Time in Relation to Percutaneous Transhepatic Portal Before 1 3 Days 6 9 Days Days Time in Relation to Percutaneous Transhepatic Portal Mean AST, ALT, LDH and γ CTP (IU) Mean WBC (cells 1 3 ) Before be because, in contrast to arterial embolization, portal vein embolization does not cause biliary ischemia or necrosis that triggers infection. This observation may suggest that portal vein embolization with ethanol can be performed without additional risk of biliary infection. Severe pain occurred in almost all patients when ethanol was injected. This is one of the disadvantages of ethanol injection. We injected lidocaine into the targeted portal 1 3 Days 6 9 Days Days Time in Relation to Percutaneous Transhepatic Portal Before 1 3 Days 6 9 Days Mean AST (IU) Mean ALT (IU) Mean LDH (IU) Mean γ GTP (IU) Days Time in Relation to Percutaneous Transhepatic Portal Fig. 4 Mean values for liver function tests obtained before and after. A, Graph shows mean WBC count. B, Graph shows mean platelet count. C, Graph shows mean serum total bilirubin level. D, Graph shows mean serum aspartate aminotransferase (AST), serum alanine aminotransferase (ALT), serum lactate dehydrogenase (LDH), and γ-glutamyl transpeptidase (γ-gtp) levels. A C vein for pain reduction, but pain control was not completely achieved. Routine administration of narcotic analgesics before ethanol injection should be considered. Elevations in total serum bilirubin and liver transaminase levels were observed after percutaneous transhepatic portal vein embolization, but the changes were transient and recovered within 1 2 weeks after percutaneous transhepatic portal vein embolization. No patient devel- B D 92 AJR:198, April 212

8 Embolization of Vein With Ethanol Injection oped fulminant liver insufficiency or died as a result of complications, and percutaneous transhepatic portal vein embolization did not have adverse effects on surgical plans. Periportal inflammatory reaction and fibrosis, such as the massive peribiliary fibrosis sometimes encountered in sclerosing cholangitis, have been reported in portal vein embolization with NBCA [7, 15, 16]. These effects were expected in portal vein embolization with ethanol because of its cytotoxicity, but there were no complications, such as inflammation-induced adhesion, that precluded subsequent surgical procedures. In this study, the frequency of complications related to the procedure was 6.%. This frequency was acceptable when compared with the reports by Kodama et al. [29] and Di Stefano et al. [24], in which the frequencies of such complications were 14.9% and 12.8%, respectively. Transient thrombocytopenia occurred and the platelet count decreased to less than cells/m 3 in three patients (2.1%). The exact underlying cause of these effects was not known, but platelet consumption in thrombosis, systemic ethanol toxicity, or toxicity by the other drugs used in the procedure were possible concerns. Complete recovery from thrombocytopenia was observed within a week. Severe thrombocytopenia requiring transfusion occurred in one patient, but the amount of ethanol injected in this case was 8 ml, which was lower than the mean value. The patient did not have any particular risk factors for thrombocytopenia. We were not able to clarify the cause of thrombocytopenia in this case and considered it as a rare complication. Contrast-enhanced CT images were obtained at 1 and 2 weeks after percutaneous transhepatic portal vein embolization. The purpose of CT at 1 week after percutaneous transhepatic portal vein embolization was to evaluate persistent occlusion of the portal veins. Although CT was used in this study, ultrasound imaging which is less expensive, easily available, and free of radiation exposure and does not require contrast agent administration can be used as a substitute for CT. Part of the increase in the size of the future liver remnant could have been due to increased blood flow as a consequence of contralateral embolization. Therefore, the period of 2 weeks plus the interval used for assessing growth might have been short. However, there have been previous reports that evaluated liver volume and function at 2 3 weeks after portal vein embolization and revealed good results [18, 22]. We evaluated not only volume change but also liver function before surgery. We do not think that the recovery of liver function could have been achieved by mere increase in blood flow. Our study has several limitations. First, this was a retrospective and monocentric study. Second, it was a nonrandomized study and lacked a direct comparison of ethanol with other embolic materials. Third, the required length of hospital stay could not be evaluated, because patients who continued to be hospitalized after portal vein embolization until surgery were included in this study. In conclusion, preoperative percutaneous transhepatic portal vein embolization with ethanol injection appears to be a feasible technique and can be expected to induce a high rate of future liver remnant hypertrophy. As an embolic material, ethanol is considered to be comparable to such other materials as spherical particles or NBCA. References 1. Kubota K, Makuuchi M, Kusaka K, et al. Measurement of liver volume and hepatic functional reserve as a guide to decision-making in resectional surgery for hepatic tumors. Hepatology 1997; 26: Mullin EJ, Metcalfe MS, Maddern GJ. How much liver resection is too much? Am J Surg 25; 19: Shimamura T, Nakajima Y, Une Y, et al. Efficacy and safety of preoperative percutaneous transhepatic portal embolization with absolute ethanol: a clinical study. Surgery 1997; 121: Kinoshita H, Sakai K, Hirohashi K, Igawa S, Yamasaki O, Kubo S. Preoperative portal vein embolization for hepatocellular carcinoma. World J Surg 1986; 1: Makuuchi M, Thai BL, Takayasu K, et al. Preoperative portal embolization to increase safety of major hepatectomy for hilar bile duct carcinoma: a preliminary report. Surgery 199; 17: de Baere T, Roche A, Vavasseur D, et al. Portal vein embolization: utility for inducing left hepatic lobe hypertrophy before surgery. Radiology 1993; 188: Imamura H, Shimada R, Kubota M, et al. Preoperative portal vein embolization: an audit of 84 patients. Hepatology 1999; 29: Azoulay D, Castaing D, Smail A, et al. Resection of nonresectable liver metastases from colorectal cancer after percutaneous portal vein embolization. Ann Surg 2; 231: Madoff DC, Hicks ME, Abdalla EK, Morris JS, Vauthey JN. Portal vein embolization with polyvinyl alcohol particles and coils in preparation for major liver resection for hepatobiliary malignancy: safety and effectiveness study in 26 patients. Radiology 23; 227: Tanaka H, Hirohashi K, Kubo S, Shuto T, Higaki I, Kinoshita H. Preoperative portal vein embolization improves prognosis after right hepatectomy for hepatocellular carcinoma in patients with impaired hepatic function. Br J Surg 2; 87: Covey AM, Tuorto S, Brody LA, et al. Safety and efficacy of preoperative portal vein embolization with polyvinyl alcohol in 58 patients with liver metastases. AJR 25; 185: Madoff DC, Abdalla EK, Gupta S, et al. Transhepatic ipsilateral right portal vein embolization extended to segment IV: improving hypertrophy and resection outcomes with spherical particles and coils. J Vasc Interv Radiol 25; 16: Abulkhir A, Limongelli P, Healey AJ, et al. Preoperative portal vein embolization for major liver resection: a meta-analysis. Ann Surg 28; 247: Vauthey JN, Chaoui A, Do KA, et al. Standardized measurement of the future liver remnant prior to extended liver resection: methodology and clinical associations. Surgery 2; 127: de Baere T, Roche A, Elias D, Lasser P, Lagrange C, Bousson V. Preoperative portal vein embolization for extension of hepatectomy indications. Hepatology 1996; 24: Madoff DC, Abdalla EK, Vauthey JN. Portal vein embolization in preparation for major hepatic resection: evolution of a new standard of care. J Vasc Interv Radiol 25; 16: Ogasawara K, Uchino J, Une Y, Fujioka Y. Selective portal vein embolization with absolute ethanol induces hepatic hypertrophy and makes more extensive hepatectomy possible. Hepatology 1996; 23: Nagino M, Kamiya J, Nishio H, Ebata T, Arai T, Nimura Y. Two hundred forty consecutive portal vein embolizations before extended hepatectomy for biliary cancer: surgical outcome and longterm follow-up. Ann Surg 26; 243: Ferrero A, Vigano L, Polastri R, et al. Postoperative liver dysfunction and future remnant liver: where is the limit? Results of a prospective study. World J Surg 27; 31: Kamiyama T, Nakanishi K, Yokoo H, et al. Perioperative management of hepatic resection toward zero mortality and morbidity: analysis of 793 consecutive cases in a single institution. J Am Coll Surg 21; 211: Denys A, Bize P, Demartines N, Deschamps F, De Baere T. Quality improvement for portal vein embolization. Cardiovasc Intervent Radiol 21; 33: Seyama Y, Kubota K, Sano K, et al. Long-term outcome of extended hemihepatectomy for hilar AJR:198, April

9 Sakuhara et al. bile duct cancer with no mortality and high survival rate. Ann Surg 23; 238: Nagino M, Nimura Y, Kamiya J, Kondo S, Kanai M. Selective percutaneous transhepatic embolization of the portal vein in preparation for extensive liver resection: the ipsilateral approach. Radiology 1996; 2: Di Stefano DR, de Baere T, Denys A, et al. Preoperative percutaneous portal vein embolization: evaluation of adverse events in 188 patients. Radiology 25; 234: Seo DD, Lee HC, Jang MK, et al. Preoperative portal vein embolization and surgical resection in patients with hepatocellular carcinoma and small future liver remnant volume: comparison with transarterial chemoembolization. Ann Surg Oncol 27; 14: Yamakado K, Takeda K, Nishide Y, et al. Portal vein embolization with steel coils and absolute ethanol: a comparative experimental study with canine liver. Hepatology 1995; 22: Denys AL, Abehsera M, Sauvanet A, Sibert A, Belghiti J, Menu Y. Failure of right portal vein ligation to induce left lobe hypertrophy due to intrahepatic portoportal collaterals: successful treatment with portal vein embolization. AJR 1999; 173: Wilms C, Mueller L, Lenk C, et al. Comparative study of portal vein embolization versus portal vein ligation for induction of hypertrophy of the future liver remnant using a mini-pig model. Ann Surg 28; 247: Kodama Y, Shimizu T, Endo H, Miyamoto N, Miyasaka K. Complications of percutaneous transhepatic portal vein embolization. J Vasc Interv Radiol 22; 13: FOR YOUR INFORMATION The American Roentgen Ray Society now provides instant Web exclusive access to its annual meeting abstracts. The abstracts, featured as a supplement to the AJR, summarize the latest comprehensive and clinically important information presented at ARRS s annual meetings. The abstracts can be viewed online by visiting AJR:198, April 212