CLINICAL INVESTIGATIONS Predisposing Factors of Bile Duct Injury After Transcatheter Arterial Chemoembolization (TACE) for Hepatic Malignancy Jeong-Sik Yu, Ki Whang Kim, Mi-Gyoung Jeong, Deok Hee Lee, Mi-Suk Park, Sang-Wook Yoon Department of Diagnostic Radiology and Research Institute of Radiological Science, Yonsei University College of Medicine, YongDong Severance Hospital, 146-92 Dogok-Dong, Gangnam-Gu, Seoul 135-270, South Korea Abstract The purpose of this study was to investigate the predisposing factors of bile duct injury after transcatheter arterial chemo- embolization (TACE) for treatment of hepatic malignancy. For patients (n = 31) with TACE-related bile duct injuries during a 36-month period, final diagnoses of the tumor, the liver profile, presence of portal vein thrombosis, total num- ber and mode of the TACE just before the development of bile duct injury were compared, respectively with those of patients without bile duct injury (n = 234) after TACE. The incidence of bile duct injury was higher in the patients with non-hepatocellular tumors than in patients with hepatocellu- lar carcinoma ( p  0.01), and higher in Child-Pugh class A patients than in B or C patients ( p  0.01). Segmental or subsegmental TACE tended to induce bile duct injury more frequently than the proximal TACE ( p = 0.01). Portal vein thrombosis, the total number of TACEs, total amount of iodized oil, and the usage of gelatin sponge were not closely related to bile duct injuries after TACE ( p  0.05). It was concluded that the chance of bile duct injury after TACE is increased in non-cirrhotic livers with good liver profile and to the more selective embolization of distal arterial branches. Key words: Bile ducts, injuries—Liver, CT—Liver, necro- sis—Liver neoplasms—Chemotherapeutic infusion Transcatheter arterial chemoembolization (TACE) has been generally used for the treatment of hepatocellular carcinoma and less frequently for other malignant tumors in the liver [1–5]. As a complication of TACE, bile duct injury has been reported intermittently since the introduction of hepatic ar- terial embolization therapy [6 –10]. Small embolic material diameter [7] and a repeated number of TACEs [10] were suggested to be related to ischemic bile duct injuries. The incidence of bile duct injury after TACE or hepatic arterial infusion chemotherapy was 12.5% in an autopsy series [9]. However, the incidence was very low (in the range of 0.9%– 2.1%) of those in some large-series follow-up CT studies [10, 11]. As a result of developments of angiographic techniques capable of high-resolution digital subtraction imaging and coaxial catheter systems capable of selecting smaller arterial branches, the concept of selective TACE to minimize hepatic parenchymal injury and preserve liver function has been generally accepted and standardized [3, 12]. However, rou- tine TACE procedures designed to embolize selectively have frequently been associated with bile duct injury and some- what unpredictably during follow-up imaging studies. To establish a safety guideline and predict the patients’ progno- sis after the procedure, the predisposing factors of the bile duct injuries, which could be a source of superimposed infection and septicemia after TACE procedures, needed clarification in the era of standardized mode of embolization therapy for hepatic malignancy. Materials and Methods From Dec. 1995 to Nov. 1998 a total of 346 consecutive patients underwent one or more sessions of TACE for the control of liver malignancy, including hepatocellular carcinoma (n = 328), chol- angiocarcinoma (n = 5), primary malignant fibrous histiocytoma (n = 1), metastases from stomach cancer (n = 4), colorectal cancer (n = 3), pancreatic cancer (n = 1), gallbladder cancer (n = 1), malignant gastrointestinal stromal tumor (n = 1), and gastrointes- tinal carcinoid tumor (n = 2). The routine protocol of TACE in our institution includes 1–20 mL of iodized oil (Lipiodol; Andre- Guerbet, Aulnay-sous-Bois, France) and 10 –50 mg of doxorubicin hydrochloride (Adriamycin: Kyowa Hakko Kogyo, Tokyo, Japan) Correspondence to: J.-Sik Yu Cardio V ascular and Interventional Radiology © Springer-Verlag New York, Inc. 2002 Cardiovasc Intervent Radiol (2002) 25:270 –274 Published Online: 3 June 2002 DOI: 10.1007/s00270-001-0049-z