obstruction unresponsive to biliary stent insertion, PDT effectively decreased serum bilirubin levels and improved quality-of-life. 5 Berr et al., 6 in the largest series reported to date, showed a similar benefit with respect to cholestasis and quality-of-life in 22 of 23 patients treated with PDT. Median sur- vival in these 2 studies was 11 months and 14 months, respectively, values comparable to a median survival of 6 to 12 months with endoscopic stent placement alone. 7,8 In these studies a thin quartz laser fiber was used that is not available in the United States. These fibers have improved flexibili- ty, which allows for their use in the biliary tract. Our aim was to develop a safe and effective method for delivering PDT within the bile duct by using slightly modified biliary catheters, guide- wires, and a commercially available laser fiber. The technical feasibility of the technique was examined in patients with complex hilar strictures and unre- sectable disease. In addition, the change was assessed in cholestatic parameters and quality-of- life indices. PATIENTS AND METHODS Over a period of 4 months, 6 patients with unresectable cholangiocarcinoma were considered for PDT (ages 38-73 years). All patients had perihilar tumors of Bismuth types IV (n = 2), III (n = 3), or II (n = 1). All but 1 patient had metastasis to regional lymph nodes. Five patients had his- tologic confirmation of malignancy and 1 patient had com- patible radiographic findings and a CA 19-9 of 110 U/mL. Experience from our institution has shown that a CA 19- 9 value greater than 100 U/mL (normal <40 U/mL) has high sensitivity and specificity for the diagnosis of cholan- giocarcinoma. 9,10 One patient had cholangiocarcinoma as a complication of primary sclerosing cholangitis and had prior combina- tion radiochemotherapy before being referred for PDT. All the patients had undergone prior endoscopic (4) or percu- taneous (2) biliary stent insertion. Endoscopic sphincter- otomy was previously performed in 3 patients. Patients received porfimer sodium (Photofrin II, Axcan, Montreal, Quebec) intravenously at a dose of 2 mg/kg 48 hours before ERCP. This compound preferentially accumu- lates in tumor cells, reaching peak tumor to normal tissue drug concentrations 24 to 48 hours after its administration. The drug requires activation with light at a wavelength of 630 nm (red light). After activation, reactive oxygen species such as singlet oxygen are generated and damage the plas- ma membrane and mitochondria of tumor cells. 11,12 These subcellular events result in vascular injury that is vital in tumor destruction. 13 Cholangiocarcinoma cell lines are extremely sensitive to this mechanism of injury. 14 All patients received 3 grams of intravenous ampi- cillin/sulbactam or an equivalent antibiotic before the pro- cedures. Endoscopic retrograde cholangiography was per- formed to define the proximal and distal extent of intraductal tumor. Tumor length was determined by using A Rumalla, T Baron, K Wang, et al. Endoscopic application of photodynamic therapy for cholangiocarcinoma 500 GASTROINTESTINAL ENDOSCOPY VOLUME 53, NO. 4, 2001 Endoscopic application of photodynamic therapy for cholangiocarcinoma Ashwin Rumalla, MD, Todd H. Baron, MD, Kenneth K.Wang, MD, Gregory J. Gores, MD, Linda M. Stadheim, RN, Piet C. de Groen, MD Background: Previous studies indicate that photody- namic therapy provides effective relief from biliary obstruction in advanced cholangiocarcinoma. This report describes a method of applying photodynamic therapy in the biliary tract by using accessories avail- able in the United States. Methods: Endoscopic retrograde cholangiography was performed to define the proximal and distal extent of intraductal tumor. Patients were injected with 2 mg/kg of sodium porfimer. Forty-eight hours later a commercially available cylindrical diffusing laser fiber was inserted into an 8F biliary catheter equipped with a 0.038 inch side-hole at its distal tip. After positioning of a 0.035 inch guidewire proximal to the biliary stricture, the preloaded catheter was advanced over the guidewire by using the monorail technique. Laser light was applied at a power of 400 mW/cm fiber for a total energy of 180 J/cm. 2 Results: Fourteen treatments were performed on 6 patients with tumors of Bismuth types IV (n = 2), III (n = 3), or II (n = 1). By using the preloaded biliary catheter, adequate positioning of the laser fiber was achieved in all patients. A fracture of the diffuser tip occurred during 1 of the treatments.Two patients developed acute cholan- gitis and 2 patients experienced skin phototoxicity. Conclusions: Photodynamic therapy for cholangiocarci- noma is safe and technically feasible with a preloaded biliary catheter and a monorail technique for catheter positioning. Photodynamic therapy (PDT) is currently being used to treat a variety of unresectable GI malignan- cies. Although approved for use in the lung and esophagus, 1 existing technology allows the applica- tion of this therapy to the stomach 2 and colon. 3,4 However, the commercially available semi-rigid cylindrical diffusing fibers have hindered the use of PDT in anatomic areas with acute angulation such as the biliary tract. Preliminary reports from Europe suggest that PDT is beneficial for patients with unresectable bil- iary tract malignancy. In 9 patients with perihilar Received June 30, 2000. For revision September 11, 2000. Accepted December 7, 2000. From the Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota. Presented as a poster during the annual Digestive Disease Week, May 21-24, 2000, San Diego, California (Gastrointest Endosc 2000;51:AB183). Reprint requests: Todd H. Baron, MD, FACP, Eisenberg 8A, 200 1st St. SW, Rochester, MN 55905. Copyright © 2001 by the American Society for Gastrointestinal Endoscopy 0016-5107/2001/$35.00 + 0 37/69/113386 doi:10.1067/mge.2001.113386