Peripancreatic carcinoma in most patients may

Original Article / Pancreas Quality of survival in patients treated for malignant biliary obstruction caused by unresectable pancreatic head cancer: surgical versus non-surgical palliation Hyung Ook Kim, Sang Il Hwang, Hungdai Kim and Jun Ho Shin Seoul, South Korea BACKGROUND: Appropriate palliation for unresectable pancreatic head cancer is most important. This study was undertaken to compare the survival of patients with biliary obstruction caused by unresectable pancreatic head cancer after surgical and non-surgical palliation. METHODS: We retrospectively reviewed 69 patients who underwent palliative treatment for unresectable pancreatic head cancer. Fifty-two patients with locally advanced disease (local vascular invasion) and 17 with distant metastatic disease were included. The patients were divided into two groups, surgical and non-surgical palliation. RESULTS: Thirty-eight patients underwent biliary bypass surgery and 31 had percutaneous transhepatic biliary drainage (PTBD). There was no significant difference in the early complications, successful biliary drainage, recurrent jaundice, and 30-day mortality between surgical palliation and PTBD. However, in 52 patients whose tumor was unresectable secondary to local vascular invasion, the rate of recurrent jaundice after successful surgical biliary palliation was lower than that in patients who had nonsurgical palliation (P<0.05). The patients who underwent surgical palliation had a longer hospital-free survival rate (P<0.001), although they had a longer postoperative hospital stay (P=0.004) during the first admission period. Author Affiliations: Department of Surgery, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, South Korea (Kim HO, Hwang SI, Kim H and Shin JH) Corresponding Author: Jun Ho Shin, MD, Department of Surgery, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, 108 Pyung-Dong, Jongno-Ku, Seoul 110-746, South Korea (Tel: 82-2-2001-2138; Fax: 82-2-2001-2131; Email: junho0521.shin@ samsung.com) 2008, Hepatobiliary Pancreat Dis Int. All rights reserved. CONCLUSIONS: In patients with preoperative evaluations showing potentially resectable tumors and/or no metastatic lesions, surgical exploration should be performed. Thus, in patients who have unresectable cancer or limited metastatic disease on exploration, surgical palliation should be performed for longer survival and better quality of survival. (Hepatobiliary Pancreat Dis Int 2008; 7: 643-648) KEY WORDS: pancreatic cancer; obstructive jaundice; palliation Introduction Peripancreatic carcinoma in most patients may be found to be unresectable at the time of diagnosis. [1] In patients diagnosed preoperatively with a resectable tumor, up to 40% of them are found to have unresectable tumor during surgical exploration. [2-4] Therefore, appropriate palliation for the main symptoms such as obstructive jaundice, duodenal obstruction, and pain is most important for patient care. [2-5] However, there is controversy over how to provide optimal palliative treatment. Obstructive jaundice can be managed by endoscopic biliary stenting (EBS), percutaneous transhepatic biliary drainage (PTBD) or surgical biliary bypass. Both retrospective and prospective randomized studies comparing EBS, PTBD, and surgical biliary bypass have shown a higher early complication rate after surgical palliation, whereas late complications appear to be more common after EBS or PTBD. [6-9] However, more recently, surgical palliation has demonstrated acceptable perioperative morbidity and mortality, with excellent long-term [2, 10] results. The aim of this retrospective review was to evaluate the efficacy of surgical biliary bypass and PTBD with Hepatobiliary Pancreat Dis Int,Vol 7,No 6 December 15,2008 www.hbpdint.com 643

Hepatobiliary & Pancreatic Diseases International or without stenting in the treatment of obstructive jaundice caused by pancreatic head cancer in terms of successful drainage, defined as adequate biliary drainage and symptomatic improvement, operative morbidity and mortality, long-term complications, and the length and quality of survival. In addition, the patients with distant metastatic disease were evaluated because of the potential bias in choosing one of the two different treatments due to relative short-term survival. Methods Study group We carried out a retrospective analysis of patients who underwent palliative treatment for unresectable pancreatic head cancer at our hospital from January 2003 to December 2006. Excluded from this review were any patients who could undergo palliative resection, had no obstructive jaundice at the time of diagnosis, had other periampullary cancer (distal bile duct, ampulla of Vater, and duodenum), and had histology other than adenocarcinoma. Sixty-nine patients were included in this study. The tumors of 52 patients were unresectable because of local vascular invasion to the superior mesenteric vein (SMV), portal vein, superior mesenteric artery (SMA), hepatic artery or celiac artery, and 17 patients had distant metastatic disease. Diagnoses were based on the findings of contrastenhanced computed tomography, magnetic resonance imaging, angiography, or exploratory laparotomy, and intraoperative ultrasonography. Thirty-two patients underwent double bypass surgery (hepaticojejunostomy or choledochojejunostomy and gastrojejunostomy), and 6 had only biliary bypass surgery (hepaticojejunostomy or choledochojejunostomy). The other 31 patients had a PTBD with or without stenting. Surgical biliary bypass and PTBD technique In patients with unresectable pancreatic head cancer, our approach consisted of 1) biliary bypass surgery with or without gastrojejunostomy after PTBD (Multi-use bile drainage catheter; Sungwon Medical, Cheongju, Korea) or 2) biliary stenting (BONASTENT TM Biliary; Standard Sci-Tech Inc., Seoul, Korea) following PTBD and duodenal stenting (BONASTENT TM Duodenal/Pyloric; Standard Sci- Tech Inc., Seoul, Korea) if clinical symptoms of gastric outlet obstruction were present. All patients had diagnostic or therapeutic PTBDs inserted before biliary bypass surgery. Bypass surgery was performed without delay to lower the bilirubin level, if patients decided to undergo surgery. We recommended all patients who did not have definite distant metastases on the preoperative studies to undergo exploratory laparotomy, since we believed that surgery was the only possible chance of curative treatment. The operative procedures employed to bypass the obstructed biliary duct and additional gastrojejunostomy were selected according to the surgeon's preference. For instance, if the blood supply of the remnant proximal bile duct was poor, hepaticojejunostomy was selected, and gastrojejunostomy was performed in patients who had a high risk for duodenal obstruction during the survival period, such as in cases with duodenal invasion. The PTBD catheter previously inserted was used for postoperative external biliary drainage until the wound of hepaticojejunostomy or choledochojejunostomy was healed without bile leakage. Conservative therapy (fasting, intravenous infusion, and antibiotic treatment) was commenced before the PTBD. In all patients, PTBDs were performed by an interventional radiologist using ultrasound and fluoroscopy guidance under local anesthesia with 1% lidocaine. A 21-gauge Chiba needle was advanced into the right or left intrahepatic duct. After placement of a guide wire and sequential dilation of the track, an 8.5F pigtail catheter was positioned with its tip in the distal bile duct. The catheter was not manipulated through the obstruction, but it was left above the obstruction and placed for external drainage. Stenting was performed, if possible, one to two weeks after PTBD to allow for internal drainage in the patients who did not have biliary bypass surgery. Cholangiography was performed immediately after placement of the pigtail catheter to confirm whether it was in a correct position within the bile duct. Data analysis The present study was relatively small, and the treatment selection was biased by several factors: accuracy of preoperative staging, patient preference for surgery, and surgeon's preference for method of treatment. However, the strengths and weakness of the two treatment policies could be evaluated. The clinical results were analyzed using a statistical analysis program package (SPSS 15.0, SPSS Inc. Chicago, IL., USA). The results were expressed as the mean±sd or median with range. The significance of the differences between the results was tested using the Chi-square test, Fisher's exact test, the Mann- Whitney test, or repeated measure ANOVA. A P value less than 0.05 was considered statistically significant. 644 Hepatobiliary Pancreat Dis Int,Vol 7,No 6 December 15,2008 www.hbpdint.com

Unresectable pancreatic head cancer Results Surgical palliation Thirty-eight patients underwent biliary bypass surgery for palliation of their malignant biliary obstruction. Their mean age was 63.6±10.0 years. The reasons for the palliative treatment were tumor invasion of the SMV or portal vein in 8 patients, SMA in 13, both in 6, hepatic artery or celiac artery in 2, hepatic metastases in 4, and peritoneal implants in 5. The operative procedures employed to bypass the biliary tract were retrocolic hepaticojejunostomy in 25 patients and retrocolic choledochojejunostomy in 13. An antecolic gastrojejunostomy was performed in 32 patients. Treatment of surgical palliation was successful in 37 (97.4%) of the 38 patients. The success was determined by normalization of the serum bilirubin levels and symptom improvement. There was one death (2.6%) due to postoperative upper gastrointestinal bleeding. The mean maximal preoperative bilirubin level was 12.8±7.4 mg/dl, and the mean minimal postoperative level was 1.2 ±0.6 mg/dl. In 7 patients (18.4%), one had upper gastrointestinal bleeding, one intraabdominal abscess, one pulmonary atelectasis with effusion, one postoperative prolonged ileus, and three wound infections. The median hospitalization after surgery was 19 days (range 1-73 days). Thirteen patients required 31 readmissions due to cholangitis, recurrent jaundice, gastric outlet obstruction, pain, or other reasons. Recurrent jaundice developed in two patients (5.4%), who underwent PTBD and stenting. One (16.7%) of 6 patients who had not received a gastrojejunostomy developed a gastric outlet obstruction requiring duodenal stent placement. The other readmitted patients were treated symptomatically. Therefore, the total hospital stay during the survival period was a median of 22 days (range 1-111 days). The median overall survival was 270 days (range 1-723 days), and the median hospital-free survival was 248.5 days (range 0-675 days). Non-surgical palliation Thirty-one patients underwent PTBD for palliation of malignant biliary obstruction. Their mean age was 64.2±12.9 years. The palliative treatment was due to tumor invasion of the SMV or portal vein (3 patients), SMA (6), both (11), hepatic artery or celiac artery (3), hepatic metastases (5), hepatic and pulmonary metastases (2), and peritoneal implants (1). The patients refused an exploratory laparotomy on their own preference. The treatment was successful in 27 (87.1%) of 31 patients. Stenting was attempted to allow internal drainage for all patients who had a successful external drainage. Stenting after PTBD was technically successful in 24 (88.9%) of the 27 patients. Two deaths (6.5%) were due to sepsis caused by aggravation of cholangitis and massive hemobilia. Four patients had persistent jaundice until death. The mean maximal pre-ptbd bilirubin level was 16.9±17.9 mg/dl, and the mean minimal post-ptbd level was 2.0±2.5 mg/dl. Two patients (6.5%) suffered from hemobilia (1) and dislodgement of the catheter (1), respectively. The median hospitalization after PTBD was 16 days (range 8-48 days). Table. Surgical versus non-surgical palliation in patients with unresectable pancreatic head cancer Surgical palliation Non-surgical palliation P value Patients 38 31 Mean age 63.6±10.0 64.2±12.9 0.836 Male 20 19 0.470 Mean pre-/post-bilirubin level (mg/dl) 12.8±7.4/1.2±0.6 16.9±17.9/2.0±2.5 0.144 Early complications (%) 7 (18.4) 2 (6.5) 0.171 Successful biliary drainage (%) 37 (97.4) 27 (87.1) 0.166 Recurrent jaundice (%) 2 (5.4) 6 (22.2) 0.061 30-day mortality (%) 1 (2.6) 2 (6.5) 0.584 Gastric outlet obstruction (%) * 0 (0.0) 4/37 (10.8) 0.118 Median postoperative hospital stay (d) 19 (range 1-73) 16 (range 8-48) 0.004 Number of readmissions 31 36 0.047 Median survival (d) 270 (range 1-723) 150 (range 21-328) <0.001 Median hospital-free survival (d) 248.5 (range 0-675) 112 (range 5-297) <0.001 *: Thirty-two patients who underwent gastrojejunostomy were compared with the other 37 patients who received non-surgical biliary palliation alone. Hepatobiliary Pancreat Dis Int,Vol 7,No 6 December 15,2008 www.hbpdint.com 645

Hepatobiliary & Pancreatic Diseases International Pancreatic head cancer Metastatic diseases on preoperative work-up Yes No Jaundice PTBD or Endoscopic biliary endoprosthesis Duodenal obstruction Endoscopic duodenal stent Widely metastatic disease Locally advanced or limited metastatic disease Surgical palliation Explorative laparotomy Pain Double bypass surgery Computed tomography or endoscopic Hepaticojejunostomy or choledocojejunostomy ultrasound guided celiac block Gastrojejunostomy, Chemical splanchnicetomy Fig. Treatment protocol for pancreatic head cancer; whether metastatic disease is present or not. Resectable disease Curative resection Twenty patients required 36 readmissions because of acute cholecystitis, cholangitis, recurrent jaundice, gastric outlet obstruction, pain, or other reasons. Recurrent jaundice occurred in 6 patients (22.2%), of whom, 4 had re-ptbd, and 2 an endoscopic endoprosthesis. Three patients (9.7%) developed gastric outlet obstruction requiring a duodenal stent placement. The other readmitted patients were treated symptomatically. Therefore, the total hospital stay during the survival period was a median of 30 days (range 8-231 days). The median overall survival was 150 days (range 21-328 days), and the median hospitalfree survival was 112 days (range 5-297 days). Statistical analysis There was no significant difference in the early complications, successful biliary drainage, recurrent jaundice, and 30-day mortality between surgical palliation and PTBD. However, the patients who underwent a surgical bypass tended to have a lower rate of recurrent jaundice, and the overall survival was longer in surgical palliation (P<0.001). Furthermore, the patients who underwent surgical palliation had a longer hospital-free survival (P<0.001), although they had a longer postoperative hospital stay (P=0.004) during the first admission period (Table). Discussion The goal of palliative operation is to alter the mode of death from that of local complications to that of systemic disease. [11] The appropriate palliative procedure should have a low morbidity and mortality, a high rate of success for relieving obstructive jaundice, and a low rate of long-term complications, thus enhancing the quality of survival. Recent studies on surgical bypass of unresectable pancreatic and periampullary cancer report operative mortality rates 1%-5% and an operative morbidity of 20%-30%. [2, 10, 12] Major advances have been made in nonoperative palliation for periampullary cancer. Percutaneous or endoscopic palliation of obstructive jaundice can provide biliary decompression with lower early morbidity compared to open biliary bypass procedures. [6-9] However, these techniques have been disappointing with regard to recurrent jaundice. In the present study, the patients who underwent a surgical bypass tended to have a lower rate of recurrent jaundice. Especially in 52 patients whose tumors were unresectable secondary to local vascular invasion, and who were expected to survive for more than several months, the rate of recurrent jaundice, after successful PTBD with or without stenting, was 30.0% (6/20) during the survival period. In contrast, the rate of recurrent jaundice after successful surgical biliary bypass was 7.1% (2/28); this difference was significant (P<0.05). In 17 patients with distant metastatic disease, there was no recurrent jaundice after successful surgical or non-surgical biliary palliation; this may be due to the short-term survival of these patients. Four duodenal obstructions (10.8%) occurred in 646 Hepatobiliary Pancreat Dis Int,Vol 7,No 6 December 15,2008 www.hbpdint.com

Unresectable pancreatic head cancer the 37 patients who received surgical or non-surgical biliary palliation alone. Lillemoe et al [3] emphasized that a retrocolic gastrojejunostomy should be performed routinely when a patient is undergoing surgical palliation for unresectable periampullary carcinoma. Another study showed clinical symptoms of gastric outlet obstruction in 5.5% of patients with a double bypass (hepaticojejunostomy and gastrojejunostomy), and in 41.4% with a single bypass (hepaticojejunostomy) for unresectable periampullary cancer. [13] In the present study, one patient who developed gastric outlet obstruction had no evidence of duodenal invasion at the time of surgery. This finding supports that a double bypass including a gastrojejunostomy is preferred to a single bypass consisting of only a biliary bypass. However, further study is needed to evaluate whether a gastrojejunostomy is needed in patients with distant metastases who are expected to survive only a few months. It has been reported that 17% of patients with pancreatic cancer who had biliary bypass alone developed duodenal obstruction after a mean of 8.6 months. [14] Gastric outlet obstruction in patients with pancreatic cancer appears to be a terminal event. Therefore, these patients should be managed to improve the quality of life. Palliative duodenal stent placement has been shown to be more beneficial than surgical gastrojejunostomy in enhancing the quality of life of patients with duodenal obstruction caused by pancreaticobiliary cancers. [15] In our study, all patients who developed gastric outlet obstruction were treated successfully with endoscopic duodenal stenting. Therefore, in the patients with distant metastases, gastric outlet obstruction may be adequately managed with endoscopic duodenal stenting. Patients with pancreatic cancer have moderate to severe pain in 30% to 40% of cases at the time of diagnosis, increasing to more than 90% shortly [16, 17] before death. The standard treatment of the pain is the administration of nonopioid analgesics, followed by weak opioids, and finally, strong opioids as necessary. [18] Most patients require increased doses of opioids during the course of the diseases. Therefore, some techniques are used to achieve a disruption of the splanchnic nerves by celiac block, since pain from pancreatic cancer is likely mediated [19, 20] via the celiac plexus. The pain of our patients was not evaluated because intraoperative chemical splanchnicectomy was not performed routinely. However, it has been reported that intraoperative chemical splanchnicectomy with alcohol significantly reduced or prevented pain in patients with unresectable pancreatic cancer and significantly improved the overall survival. [16] Hence, we consider the use of chemical splanchnicectomy in all patients with preoperative pain at the time of palliative surgery. The positive effects of surgical biliary bypass on long-term complications and the quality of survival could not be confirmed in the present study. Surgical biliary bypass appears to have a positive effect on hospital-free and overall survival. However, this effect must be interpreted with caution because of the selection bias inherent in this study. The patients who did not receive explorative laparotomy and surgical palliation may already have had limited metastatic diseases not detected on the preoperative work-up. Actually, surgical palliation was performed only in patients who did not have metastatic disease on the preoperative imaging studies. Thus, longterm survival and a good quality of survival could be achieved in these patients. In conclusion, nonoperative palliative treatments are increasingly employed in patients with unresectable pancreatic head cancer, reducing the number of patients undergoing surgical palliation. In patients whose preoperative work-up confirms metastatic lesions, management should be performed to best suit the individual patient's clinical signs and symptoms, prognosis, and overall medical condition. However, in patients with preoperative evaluations showing potentially resectable tumors and/or no metastatic lesions, surgical exploration should be performed because of the chance to confirm the nature and full extent of the tumor and the potential for a resection to cure. And then, if it is concluded to be unresectable or to have limited metastatic disease on surgical exploration, surgical biliary bypass and gastrojejunostomy should be performed (Fig.) for longer survival and better quality of survival. Funding: None. Ethical approval: Not needed. Contributors: SJH proposed the study, KHO wrote the first draft. KHO, HSI, and KH analyzed the data. All authors contributed to the design and interpretation of the study and to further drafts. SJH is the guarantor. Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References 1 Kelly DM, Benjamin IS. Pancreatic carcinoma. Ann Oncol 1995;6:19-28. 2 Sohn TA, Lillemoe KD, Cameron JL, Huang JJ, Pitt HA, Hepatobiliary Pancreat Dis Int,Vol 7,No 6 December 15,2008 www.hbpdint.com 647

Hepatobiliary & Pancreatic Diseases International Yeo CJ. Surgical palliation of unresectable periampullary adenocarcinoma in the 1990s. J Am Coll Surg 1999;188:658-669. 3 Lillemoe KD, Cameron JL, Hardacre JM, Sohn TA, Sauter PK, Coleman J, et al. Is prophylactic gastrojejunostomy indicated for unresectable periampullary cancer? A prospective randomized trial. Ann Surg 1999;230:322-330. 4 van Wagensveld BA, Coene PP, van Gulik TM, Rauws EA, Obertop H, Gouma DJ. Outcome of palliative biliary and gastric bypass surgery for pancreatic head carcinoma in 126 patients. Br J Surg 1997;84:1402-1406. 5 van Heek NT, van Geenen RC, Busch OR, Gouma DJ. Palliative treatment in "peri"-pancreatic carcinoma: stenting or surgical therapy? Acta Gastroenterol Belg 2002; 65:171-175. 6 Shepherd HA, Royle G, Ross AP, Diba A, Arthur M, Colin-Jones D. Endoscopic biliary endoprosthesis in the palliation of malignant obstruction of the distal common bile duct: a randomized trial. Br J Surg 1988;75:1166-1168. 7 Smith AC, Dowsett JF, Russell RC, Hatfield AR, Cotton PB. Randomised trial of endoscopic stenting versus surgical bypass in malignant low bileduct obstruction. Lancet 1994;344:1655-1660. 8 van den Bosch RP, van der Schelling GP, Klinkenbijl JH, Mulder PG, van Blankenstein M, Jeekel J. Guidelines for the application of surgery and endoprostheses in the palliation of obstructive jaundice in advanced cancer of the pancreas. Ann Surg 1994;219:18-24. 9 Bornman PC, Harries-Jones EP, Tobias R, Van Stiegmann G, Terblanche J. Prospective controlled trial of transhepatic biliary endoprosthesis versus bypass surgery for incurable carcinoma of head of pancreas. Lancet 1986;1:69-71. 10 Lesurtel M, Dehni N, Tiret E, Parc R, Paye F. Palliative surgery for unresectable pancreatic and periampullary cancer: a reappraisal. J Gastrointest Surg 2006;10:286-291. 11 Potts JR 3rd, Broughan TA, Hermann RE. Palliative operations for pancreatic carcinoma. Am J Surg 1990;159: 72-78. 12 Singh S, Sachdev AK, Chaudhary A, Agarwal AK. Palliative surgical bypass for unresectable periampullary carcinoma. Hepatobiliary Pancreat Dis Int 2008;7:308-312. 13 Van Heek NT, De Castro SM, van Eijck CH, van Geenen RC, Hesselink EJ, Breslau PJ, et al. The need for a prophylactic gastrojejunostomy for unresectable periampullary cancer: a prospective randomized multicenter trial with special focus on assessment of quality of life. Ann Surg 2003;238:894-905. 14 Watanapa P, Williamson RC. Surgical palliation for pancreatic cancer: developments during the past two decades. Br J Surg 1992;79:8-20. 15 Maetani I, Tada T, Ukita T, Inoue H, Sakai Y, Nagao J. Comparison of duodenal stent placement with surgical gastrojejunostomy for palliation in patients with duodenal obstructions caused by pancreaticobiliary malignancies. Endoscopy 2004;36:73-78. 16 Lillemoe KD, Cameron JL, Kaufman HS, Yeo CJ, Pitt HA, Sauter PK. Chemical splanchnicectomy in patients with unresectable pancreatic cancer. A prospective randomized trial. Ann Surg 1993;217:447-457. 17 Yeo CJ. Pancreatic cancer: 1998 update. J Am Coll Surg 1998;187:429-442. 18 Carr DB, Goudas LC, Balk EM, Bloch R, Ioannidis JP, Lau J. Evidence report on the treatment of pain in cancer patients. J Natl Cancer Inst Monogr 2004:23-31. 19 Gress F, Schmitt C, Sherman S, Ikenberry S, Lehman G. A prospective randomized comparison of endoscopic ultrasound- and computed tomography-guided celiac plexus block for managing chronic pancreatitis pain. Am J Gastroenterol 1999;94:900-905. 20 Pietrabissa A, Vistoli F, Carobbi A, Boggi U, Bisà M, Mosca F. Thoracoscopic splanchnicectomy for pain relief in unresectable pancreatic cancer. Arch Surg 2000;135:332-335. Received June 30, 2008 Accepted after revision September 2, 2008 648 Hepatobiliary Pancreat Dis Int,Vol 7,No 6 December 15,2008 www.hbpdint.com