Endoscopic and Operative Palliation Strategies for Pancreatic Ductal Adenocarcinoma Alexander Stark and O. Joe Hines Malignant biliary obstruction, duodenal, and gastric outlet obstruction, and tumor-related pain are the complications of unresectable pancreatic adenocarcinoma that most frequently require palliative intervention. Surgery involving biliary bypass with or without gastrojejunostomy was once the mainstay of treatment in these patients. However, advances in non-operative techniques—most notably the widespread availability of endoscopic biliary and duodenal stents—have shifted the paradigm of treatment away from traditional surgical management. Questions regarding the efficacy and durability of endoscopic stents for biliary and gastric outlet obstruction are reviewed and demonstrate high rates of therapeutic success, low rates of morbidity, and decreased cost. Surgery remains an effective treatment modality, and still produces the most durable relief in appropriately selected patients. Semin Oncol 42:163-176 & 2015 Elsevier Inc. All rights reserved.

T

he incidence of pancreatic cancer in the United States is rising, 48,960 46,420 new cases and 40,560 pancreatic cancer-related deaths in 2014.1 Despite advances in diagnostic imaging, pre-operative diagnosis, surgical management, and the rise of specialized centers in the treatment of pancreatic adenocarcinoma, the prognosis for the majority of these patients remains strikingly poor.2 Consequently, the management of these patients is overwhelmingly one of palliation, directed at relieving symptoms and improving quality of life. Complete R0 surgical resection remains the only chance for cure; however, 80%–90% of patients present with evidence of metastatic or locally advanced disease precluding curative resection.3 The most common complications of advanced pancreatic cancer requiring intervention are malignant biliary obstruction, duodenal or gastric outlet obstruction, and tumor-associated pain—related to direct extension of the tumor into the bile duct, duodenum, and celiac plexus, respectively. Biliary obstruction and gastric outlet obstruction predominantly occur in

Department of Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA. Conflicts of interest: Address correspondence to O. Joe Hines, MD, Department of Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA 90095. E-mail: [email protected] 0093-7754/ - see front matter & 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1053/j.seminoncol.2014.12.014

patients with lesions of the head of the pancreas, the most common location for pancreatic adenocarcinoma. Indeed, despite typically presenting earlier than lesions in the body or tail of the pancreas, a recent series indicate that lesions of the head of the pancreas make up approximately 70% of unresectable pancreatic adenocarcinomas.4 Open surgical bypass for biliary obstruction and open gastric bypass for duodenal or gastric outlet obstruction have historically been the mainstays of surgical palliation for unresectable disease. Simultaneous biliary and gastric bypass, known as “double loop bypass”, has been used as a prophylactic operation. Opioid analgesics have been used for chronic pain. Advances in endoscopic retrograde cholangiopancreatography (ERCP), endoscopic ultrasound (EUS), and percutaneous computed tomography (CT)-guided techniques have provided alternatives to surgery. Recent series confirm that while the number of resections for curative intent has remained stable over time, the number of patients undergoing operative palliation has significantly decreased.5 In fact, most centers now preferentially palliate biliary and gastric outlet obstruction with endoscopically placed biliary and duodenal stents. These non-operative techniques have been proven effective and allow the patient to quickly receive or return to treatment with chemotherapy and/or radiotherapy as directed by the medical oncologist. The most appropriate course of management can be debated in many cases. The durability of nonoperative approaches has been a historical concern,

Seminars in Oncology, Vol 42, No 1, February 2015, pp 163-176

163

164

but many clinical trials—retrospective and prospective randomized—have addressed this issue. What’s more, despite advances in preoperative staging, up to a third of patients taken to surgery are found to be unresectable at the time of laparotomy. These patients may be symptomatic at the time of laparotomy and in need of palliation for jaundice, pain, and gastric outlet obstruction.6 While still in the operating room, is the surgeon justified in performing operative bypass? Should the surgeon perform prophylactic bypass in asymptomatic patients? Or should the surgeon close the abdomen without performing a palliative procedure and defer all initial palliative interventions to the endoscopist? This review will present the historical discussion and current data supporting the approaches to palliation of pancreatic ductal adenocarcinoma.

BILIARY OBSTRUCTION Biliary obstruction of the intrapancreatic portion of the common bile duct as a result of malignant disease in the head of the pancreas is the most common reason for palliative surgical intervention in patients with unresectable pancreatic adenocarcinoma. Historically, jaundice is seen in approximately 48-57% of patients with unresectable disease. Painless jaundice—one of the few symptoms of pancreatic adenocarcinoma that allows for “early” detection of disease—is uncommon in patients with advanced disease, as these patients are significantly more likely to present with concomitant abdominal pain, back pain, nausea, and vomiting.6,7 Newer series indicate that biliary obstruction is the presenting symptom in 51%–72% of unresectable patients, while developing in as many as 81% during the course of disease if not intervened.4,5 The clinical constellation caused by malignant biliary obstruction begins with marked elevation in serum bilirubin. Jaundice ensues and is accompanied by pruritus in some 20% of patients, which is difficult to treat with medical therapy alone. Fat malabsorption occurs as a result of decreased secretion of bile acids, contributing to malnutrition and cachexia. Finally, unrelieved biliary obstruction ultimately leads to high risk of cholangitis, cholestasis, hepatic dysfunction, and liver failure. In one series, up to 38% of patients that did not undergo palliative therapy died of these terminal sequelae.7 Indeed, early data suggested that biliary bypass prolonged survival,8 although that notion has been repeatedly challenged. Nevertheless, relief of biliary obstruction remains necessary to maintain quality of life and to prevent the rapid deterioration in clinical condition that accompanies eventual liver failure. Operative biliary bypass has been the historical gold standard for palliation of malignant biliary

A. Stark and O.J. Hines

obstruction, and has been considered both effective and durable. Therapeutic failure—defined as the rate of recurrent jaundice—is often reported to be as low as 2%–5%.5,6,9 This durability has been challenged, with some series finding the rate of recurrent jaundice to be in the range of 11%–15%.7,10 Furthermore, major morbidity and mortality have been associated with operative palliation. For these reasons, endoscopic biliary stent placement in many places has become the intervention of choice for palliation of obstructive jaundice.

Surgical Therapy for Biliary Obstruction Conventional surgical management of the obstructed distal bile duct has been performed as an open operation consisting of a cholecystoenterostomy, choledochoenterostomy, or hepaticojejunostomy—the choice of which is as frequently determined by surgeon preference as by extent of disease. Either the duodenum or the jejunum may be used as the conduit, with the jejunum being the more common site of enteric anastomosis given relative technical ease and theoretical concerns for eventual tumor extension involving the duodenum. Finally, the choledocho-, cholecysto-, or hepaticojejunostomy may be performed either as loop or a Roux-en-Y reconstruction. Loop construction avoids a jejunojejunostomy and is quicker; however, expert opinion maintains that Roux-en-Y is preferred in patients with anticipated long-term survival, as it lowers the subsequent risk of cholangitis.7 Regarding the technique of biliary bypass, hepaticojejunostomy is reserved for cases in which either the extent of bile duct involvement precludes choledochojejunostomy, or the cystic duct-common duct confluence is compromised precluding cholecystojejunostomy; in all other cases, the decision to use the bile duct or the gallbladder is to this day controversial. All of the above types of biliaryenteric bypasses have been demonstrated to be efficacious in retrospective series. One of the earliest studies published on this topic included a review of 8,000 patients from 1965–1980, in which the authors concluded that a loop cholecystojejunostomy should be performed in all patients unless the cystic duct-common duct confluence was directly involved.8 Data from the UCLA experience through 1990 demonstrated equal efficacy in terms of relief of biliary obstruction between cholecystojejunostomy and choledochojejunostomy; however, use of the bile duct was associated with significantly higher rate of morbidity (38.3% v 27%) and mortality (11.7% v 5.4%).7 Conversely, a contemporaneous series out of France found that cholecystoenteric bypass was associated with a higher post-operative mortality (20% v 14%), lower long-term morbidity

Endoscopic and operative palliation strategies

(26% v 35%), and a lower survival rate (means: 3.2 v 5.2 months) in comparison to choledochoenteric bypass. In that series, using the duodenum or the jejunum made no difference on postoperative mortality, morbidity, incidence of recurrent jaundice (8% v 7%), or median survival.11 Data from randomized clinical trials support the conclusion of the French retrospective series, suggesting that choledochoenterostomy is superior to cholecystoenterostomy when both options are technically feasible. Sarfeh et al studied these techniques in a randomized setting of choledochoenterostomy versus cholecystoenterostomy for non-calculous obstruction (malignant or chronic pancreatitis). Despite being a small study, they found a significantly higher percentage of therapeutic failure after cholecystoenterostomy, especially in those with malignant obstruction.12 Rosemurgy et al simultaneously found that use of the bile duct achieved more reliable relief of jaundice. Where choledochoenterostomy was effective for 93% of patients in the short-term and 83% of patients in the long-term, cholecystoenterostomy was successful only 50% of the time—both in the short- and long-term. In this trial perioperative morbidity and mortality were alarmingly high, approaching 60% and 23%, respectively, for cholecystoenteric bypass and nearly as high for choledochoenteric bypass.13

Endoscopic Biliary Drainage Since operative biliary bypass can be associated with high morbidity, endoscopic techniques have advanced as an alternative. The procedure involves sedation without general anesthesia, endoscopic visualization and cannulation of the major papilla, and then cholangiopancreatography to define the area of obstruction or stenosis. A stent is then deployed across the lesion, and sphincterotomy is generally performed as an adjunct. In cases in which cannulation of the papilla is difficult or passage of a wire past the lesion is difficult to achieve endoscopically, the so-called “rendezvous” technique may be employed, in which a percutaneous transhepatic biliary catheter is placed prior to endoscopy and used as a guide for crossing the lesion with a wire prior to stent placement. The stent deployed may be made of plastic or metal. Plastic stents (typically 7-F or 10-F in size) are easily retrieved, and may be totally removed endoscopically or during surgery. Metal stents (SEMS) are comprised of a selfexpanding metal mesh that may either be uncovered or covered with a thin, non-porous membrane. These stents are generally not retrievable, endoscopically. Successful placement of a biliary stent in patients with malignant biliary obstruction is generally possible in greater than 90% of patients. Inability to place a stent is most often related to difficulty in cannulating the papilla, or hilar infiltration by the tumor.4

165

Early complications are related to the procedure itself and are typically unrelated to the type of stent deployed. Pancreatitis, bleeding, and infection (ie, cholangitis) constitute the majority of these early complications and occur in approximately 5% of patients. Intestinal perforation that is clinically relevant or requiring intervention appears to be rare.14 An early, small randomized trial showed a higher (but statistically insignificant) rate of infectious complications including cholangitis, abscess, and empyema of the gallbladder when compared with traditional surgical bypass (36% v 20%).15 This rate has decreased as experience has led to improved technique. Finally, the theoretical risk of cholecystitis caused by a plastic or covered metal stent obstructing the cystic duct has not significantly materialized,16,17 although most practitioners recommend stenting below the cystic duct-common hepatic duct confluence whenever technically possible. Late complications depend on stent type and typically are a result of stent dysfunction in the form of lumen obstruction or stent migration. This typically results in recurrent jaundice, occurring in early trials in as many as 17%–38% of patients.15,18,19 Plastic stents obstruct commonly as a result of a combination of biliary sludge and bacterial overgrowth. Measures to prevent build up of bacteria with prophylactic antibiotic administration have been ineffective,17,20 and plastic stents must be exchanged at regular intervals for this reason. SEMS obstruct less commonly by bacterial overgrowth and biliary sludge, but are more frequently obstructed by tumor ingrowth, tumor overgrowth, or sludge.17 The development of covered SEMS was undertaken to combat the propensity for tumor ingrowth, and has been efficacious in that regard. The risk of stent migration depends on the type of stent. Fully covered SEMS have the highest rate of migration, occurring in up to 20% of patients. Five percent of plastic stents and partially covered SEMS migrate, and only 1% of uncovered SEMS migrate.14 Newer covered SEMS may be made with an antimigration mechanism consisting of uncovered, flared ends. These SEMS theoretically combine the anti-tumor ingrowth properties of covered stents with the antimigratory properties of uncovered stents. Indeed, a recent randomized trial comparing traditional, uncovered SEMS with the “anti-migratory,” covered SEMS found that the overall risk of stent dysfunction was significantly reduced.21 Of the plastic stents, polyethylene stents are preferable to Teflon due to better performance and lower risk of migration.22

Plastic Versus Self-Expanding Metal Stents When first introduced, most biliary stents placed were plastic. With the advent of SEMS purported to

166

have significantly longer patency, many studies— both retrospective and randomized—endeavored to compare the two types of stents with regard to efficacy, safety, and cost. In all studies, both types of stents are deployed with high rates of success and achievement of initial decompression of the biliary tract.4,16 Overall, in patients with expected survival greater than 4 months, SEMS have been found to be preferable to plastic stents by way of lower stent failure, decreased total hospitalizations secondary to stent-related issues, a lower risk of cholangitis, and overall lower costs.23 A 2006 Cochrane review of nine studies showed that metal stents are associated with a lower risk of recurrent biliary obstruction (RR 0.48). No difference was found with regard to technical or therapeutic failure, complications, or 30-day mortality.22 In general, stent dysfunction is seen in plastic stents at a rate twice that observed with metal stents.14 With regard to overall primary patency, plastic stents have a reported median patency ranging from between 1.8–4.2 months, in comparison to 3.6–9.1 months for SEMS.4,17,24,25 One retrospective review found the median primary patency of covered SEMS to be nearly 11 months.16 Although these ranges vary from study to study, direct head-to-head comparisons have uniformly favored SEMS in this regard. One such study found that SEMS have as much as five times longer primary patency than plastic stents. 24 Whereas obstruction of a plastic stent is treated by removal and placement of a new stent, SEMS obstruction can be treated with a second stent in a “stent-in-stent” fashion. This was studied retrospectively, and determined that median patency for secondary metal stents was superior to that with secondary plastic stents. Using at least one covered stent had the longest secondary median patency.26 Plastic stents are themselves substantially cheaper than SEMS (with reported difference in price on the order of magnitude of $20 for plastic and $900 for SEMS).25 However, after factoring in the costs necessary for additional ERCP and hospitalizations required for both the routine and therapeutic exchange of plastic stents, multiple trials have found that overall costs are lower after initial treatment with SEMS.17,23,25 One study associated the initial use of SEMS with a 28% decrease in overall procedures when compared to initial use of plastic stents.25 The time period at which the additional costs associated with treating stent dysfunction overcome the initial expense of the SEMS appears to be at approximately 4 months.14 Thus the cost supremacy of SEMS is predicated on two factors: patient survival greater than 4 months, and avoidance of early SEMS dysfunction requiring intervention. Hamada et al performed a review targeting early (o3 months) SEMS dysfunction, finding that

A. Stark and O.J. Hines

duodenal invasion was the only significant risk factor for such early failure.27 Finally, in the case of borderline unresectable patients undergoing neoadjuvant therapy in hopes of eventually becoming surgical candidates, SEMS have been found superior to plastic stents for all the reasons described above. In addition, the use of plastic stents in patients receiving neoadjuvant therapy has been associated with a seven times higher rate of complication and three times higher rate of hospitalization when compared to SEMS.24 Furthermore, a randomized study of patients undergoing resection for cure found that routine preoperative biliary drainage with plastic stents was associated with a significant increase in serious complications.28 Thus the role for plastic stents in the management of malignant biliary obstruction has become quite limited, indicated only in patients with metastatic disease if survival is expected to be less than four months and the patient is definitively excluded from candidacy for neoadjuvant therapy or future operative intervention.17 SEMS should be used for the majority of patients.

Surgical Biliary Bypass Versus Endoscopic Biliary Stent Early retrospective studies comparing surgical palliation of malignant biliary obstruction to nonoperative approaches (both endoscopic and percutaneous) suggested therapeutic equivalency in survival, morbidity, and efficacy, along with significantly lower costs associated with non-operative palliation.29 A more recent retrospective series reported similar overall morbidity and mortality with surgical palliation and endoscopic biliary stents30; however, the lack of randomization allows for significant patient selection bias. A Mayo Clinic study retrospectively compared the cost of surgical versus endoscopic palliation, noting that not only were initial hospitalization charges higher in the surgical group, but subsequent admissions carried a higher average cost as well, despite the need for routine stent exchanges. No differences in survival were noted.31 The earliest randomized trials comparing surgical biliary bypass to percutaneous or endoscopic biliary stents echoed the retrospective studies mentioned above, finding no difference in overall survival, but uniformly shorter initial hospitalizations associated with stent placement.15,18,19,32,33 Whereas the trial concerning percutaneous palliation found that early benefits in decreased hospitalization were offset by late admissions for stent obstruction and gastric outlet obstruction,18 trials of endoscopy versus surgery found that total length of hospitalization was shorter nonetheless.19 Indicative of these trials, a randomized controlled trial (RCT) published in The

Endoscopic and operative palliation strategies

167

Table 1. Summary of Randomized Data Comparing Surgery and Endoscopic Stent for Palliation of Malignant Biliary Obstruction Shepherd et al,19 1988

Andersen et al,15 1989

Surgery

Surgery

Stent

Stent

Smith et al,32 1994 Surgery

Stent

Artifon et al,35 2006 Surgery

Stent

No. randomized No. treated Treatment type/pts

25 27 25 25 103 101 15 15 25 23 19 25 101 100 15 15 CCJ/13 Plastic/23 CCJ/13 Plastic/25 CCJ/30 Plastic/100 CCJ/2 SEMS/15 CCD/10 CDJ/3 CDJ/28 CDJ/13 CDD/3 CDD/36 Technical success (%) 92 91 84 96 94 95 100 100 Recurrent obstruction / 0 43 84 n/a 2 36 0 27 initial stent failure (%) Complication rate (%) 40 22 20 36 29 11 (P ¼.02) 60 40 30-day mortality (%) 20 9 24 20 14 3 ( P ¼.01) 0 0 Median total 13 8 27 26 20 26 ( P ¼.001) 9 2 hospitalization (d) Median survival (d) 126 154 100 84 182 147 130 90 All available statistically significant P values shown. Abbreviations: pts, patients; CCJ, cholecystojejunostomy; CDJ, choledochojejunostomy; CDD, choledochoduodenostomy; SEMS, selfexpanding metal stents.

Lancet 1994 found decreased initial procedural mortality (3% v 14%), complications (11% v 29%), and shorter hospital stay at the price of much higher recurrent jaundice (36% v 2%) in patients treated with plastic stents.32 These early RCTs must be interpreted in light of the fact that they were performed prior to widespread availability of SEMS and prior to common use of duodenal stents for malignant gastric outlet obstruction. More recent trials comparing surgery to SEMS find again that endoscopic palliation is associated with decreased cost, shorter hospitalization, but increased late complications predominantly in the form of recurrent jaundice.34,35 In one study, late complications were equivalent after considering interventions necessary to treat wound dehiscence and surgical site infection. In that group, quality of life was subjectively higher in the endoscopy group.35 These RCTs are summarized in Table 1. The pooled results of RCTs have been consistent, with a Cochrane review and 2006 meta-analysis finding a reduced risk of complication (RR 0.6) but a higher risk of recurrent biliary obstruction prior to death (RR 18.59). Technical success is high and similar between the two procedures. No definitive advantage in quality of life between surgical or endoscopic therapy can be concluded.22,36 A more recent meta-analysis of RCTs with a total of 379 patients reported similar results.37 Thus endoscopic biliary stents are shown to be as effective as surgical

bypass in alleviating malignant biliary obstruction, but with less procedural morbidity; the durability of the initial procedure is significantly lower, but easily treated by repeat endoscopy and re-stenting. A comprehensive algorithm for the treatment of malignant biliary obstruction is presented in Figure 1.

DUODENAL AND GASTRIC OUTLET OBSTRUCTION Invasion of the duodenum leading to duodenal and gastric outlet obstruction (GOO) is a serious event in the progression of adenocarcinoma of the head of the pancreas. Patients experience nausea and vomiting that becomes intractable, and oral intake eventually becomes impossible or markedly reduced. It has been dubbed a “terminal event” in the local progression of pancreatic cancer, for when left untreated, malnutrition and cachexia along with a markedly decreased quality of life contribute strongly to the clinical deterioration of affected patients.38 Although nausea and vomiting will occur in 30%–50% of patients, GOO caused by duodenal invasion is generally considered to occur in 10%–25% of patients with unresectable disease.4,7,8,39 For 6% of patients, GOO is the presenting symptom.4 Finally, in those patients that undergo operation without gastrojejunostomy during the initial procedure, surgical series show that 10%–25% will have symptoms related to GOO or require intervention for GOO prior to death.7,8,40

168

A. Stark and O.J. Hines

PDAC with clinical or laboratory evidence of biliary obstrucon

Yes

Surgical exploraon

Potenally Resectable?

Expected survival >4 months?

No

Yes

No

Yes

Sll resectable?

Endoscopic palliaon with SEMS

Resecon

Endoscopic palliaon with plasc stent

No

Impending duodenal obstrucon or preoperave nausea/voming

Yes

Stent dysfuncon or failure?

No Technical failure

Double loop bypass

Aempt repeat endoscopic management prior to surgery

Biliary bypass

Figure 1. Comprehensive algorithm for surgical and endoscopic management of malignant biliary obstruction secondary to pancreatic adenocarcinoma.

The historic and gold standard treatment for malignant GOO is open surgical gastrojejunostomy. Early series showed poor outcomes in unresectable patients after palliative gastrojejunostomy; one study found that 40% of those without GOO, 70% of those found to have duodenal invasion upon exploration, and 90% of those with preoperative symptoms of GOO died within 30 days of surgery.38 Thankfully, in specialized centers palliative gastrojejunostomy has become much less morbid. Gastrojejunostomy has been advocated frequently as a prophylactic procedure in those patients found to have unresectable disease at the time of exploration. However, this paradigm is once again shifting away from routine gastrojejunostomy with the development of endoscopically placed duodenal stents.

Surgical Gastrojejunostomy—Open and Laparoscopic Open gastrojejunostomy may be performed alone or in combination with biliary bypass. The anastomosis may be hand-sewn or stapled, according to the surgeon’s preference. Although some surgeons advocate an antecolic gastrojejunostomy to reduce the risk

of involvement by tumor invasion of the transverse mesocolon,7 a retrocolic anastomosis has been used frequently without significant complications 39,41 The feasibility of laparoscopic gastrojejunostomy was initially established in a number of small retrospective studies. At UCLA, eight of nine patients with unresectable pancreatic cancer complicated by GOO were successfully palliated with laparoscopic gastrojejunostomy; no conversions to open procedures were required. Two of these patients had prior laparotomies and 4 had failed duodenal stenting.42 In a similarly sized review, Alam et al found similarly high rates of successful palliation in patients with malignant GOO.43 Direct comparison of laparoscopic versus open gastrojejunostomy has been performed, although data are limited. Navarra et al randomized 24 patients with malignant GOO to laparoscopic versus open gastrojejunostomy, finding that the laparoscopic approach was associated with a significant decrease in time to oral intake (4.08 v 6.25 days). There was a non-significant trend towards decreased incidence of delayed gastric emptying and decreased length of stay in the laparoscopic cohort.44 Laparoscopic gastrojejunostomy is therefore a viable, less invasive option for surgical palliation of GOO.

Endoscopic and operative palliation strategies

Endoscopic Therapy for Duodenal and Gastric Outlet Obstruction Duodenal stenting is a newer, but now wellestablished technique for the palliation of malignant GOO in which a large (20–22 mm diameter, 60–90 mm length) self-expanding metal stent is deployed across the stricture after endoscopic positioning with or without fluoroscopic guidance. Stents may be covered or uncovered. Initial small series not only demonstrated adequate safety and efficacy, but also suggested substantially improved time to oral intake —as fast as within 24 hours of the procedure.45 This technique has since been adopted rapidly by many centers as a result of the promise of an early return to oral intake without the morbidity associated with surgical gastrojejunostomy. The most serious complications are related to the procedure itself, most prominently intestinal perforation.46 Larger retrospective series of endoscopic duodenal stents in malignant GOO (not limited to pancreatic cancer) find efficacy rates as high as 88%–91% in achieving improvement in oral intake. Individual studies tend to be limited in their ability to determine complications secondary to limited sample size.4,47 A systematic review of similar case series inclusive of 606 patients analyzed in an intention-totreat method reported an overall 97% technical and 87% clinical success rate. Overall mean time to clinical improvement was 3.7 days. There were no procedure-related mortalities, and only 1.2% experienced perforation or bleeding. The overall complication rate was 26.7%, with stent obstruction and migration accounting for the vast majority.48 It must be noted that overall survival duration was a mere 12 weeks, and that even more stent-related complications would be expected given longer survival. In a prospective trial of 51 patients with a confirmed diagnosis of GOO secondary to pancreatic adenocarcinoma, similar results were found (98% technical and 84% clinical success). Despite short survival (average 10.2 weeks), 12 patients developed late stent obstruction; the majority were treated with a second duodenal stent. There were six major complications related to stent placement (11%), five of which were lethal in patients with poor overall medical condition.49 Finally, despite an increased risk of tumor ingrowth and resultant obstruction, it appears that in patients with poor overall survival uncovered duodenal stents may be preferable to covered stents as they are associated with a decreased risk of migration.46

Surgical Gastrojejunostomy Versus Endoscopic Duodenal Stent In patients with duodenal obstruction or GOO secondary to pancreatic adenocarcinoma, the

169

decision to perform surgical bypass versus endoscopic palliation is one predicated on expected survival, functional status, and local expertise. Whereas recurrent GOO secondary to stent obstruction can occur in approximately 25% of those palliated endoscopically, the incidence of recurrent GOO in patients undergoing surgical gastrojejunostomy is as low as 2% in large retrospective series.6 However, the physical stress and morbidity associated with surgery has been theorized to contribute to decreased quality of life and even to decreased survival, as reported in earlier comparative case series.50 Retrospective series, randomized trials, and metaanalyses uniformly find that endoscopic palliation for GOO is associated with decreased initial length of hospitalization and quicker initial return to oral intake (2–5 v 6–8 days).50–54 In contrast to early small retrospective series, no difference in mortality has been found in more recent data.53,54 Increased early minor complications associated with surgery appear to be offset by late stent obstruction requiring re-intervention.53,55 Finally, the SUSTENT multicenter randomized trial in 2010 found that although endoscopic palliation was associated with a significantly quicker return to eating (5 v 8 days), the relative percentage of patients tolerating a diet in the endoscopic group decreased after 30 days, and at 60 days the surgical cohort had significantly better oral intake. After adjustment for survival, surgery was associated with more total days tolerating a diet at the expense of overall total cost.55 In a small, randomized trial comparing endoscopy specifically to laparoscopic gastrojejunostomy, surgery was associated with significantly higher initial post-procedure pain and subjectively decreased physical health at 1 month. No data were collected regarding long-term efficacy.56 Table 2 summarizes the results of these randomized trials. As a result of the above studies, it appears that the durability of surgical gastrojejunostomy is superior to endoscopic palliation. Thus predicting the length of survival after treatment for malignant GOO has become important in guiding the choice of therapy. A large retrospective series was so analyzed by the authors of the SUSTENT study, with multivariate analysis finding only preoperative performance status as predictive of survival after intervention for malignant GOO. Those with World Health Organization score (WHO) o2 (fully ambulatory) had median survival of 139 days and surgical bypass is recommended; those with WHO scores 42 (450% of time spent in bed) had median survival of 31 days, favoring endoscopic palliation. Those with WHO scores of 2 (o50% of time spent in bed, but symptomatic) had intermediate survival, and therefore must be decided on a case-by-case basis.57

170 Table 2. Summary of Randomized Data Comparing Surgery and Duodenal SEMS for Palliation of Malignant Duodenal or Gastric Outlet Obstruction Fiori et al,54 2004

Mehta et al,56 2006

Jeurnink et al,55 2010

Study

Surgery

Stent

Surgery

Stent

Surgery

Stent

No. randomized No. treated Technique/pts

9 9 OGJJ/9

9 9 Endoscopic, covered/9

14 13 LGJJ/13

13 12 Endoscopica/1

18 17 OGJJ/15

21 20 Endoscopica/20

100 62

Fluoroscopica/11 83 0

LGJJ/2 83 0b 33c

86 29 (P ¼ .02) 19

23 11.4 0 – –

17 5.2 (P ¼ .02) 0 – –

–

–

Technical success (%) Complication rate (%)

100 22

22

30-day mortality (%) 0 0 Initial length of stay (d) 10 3.1 (P o.0001) Re-intervention (%) 11 11 Time to oral intake (d) 6.3 2.1 (P o.0001) Long-term relief No difference in oral – intake after 15d Total cost ($) – –

15 7 (P ¼ .04) 2 10 (P o.01) 8 5 (P ¼.01) Surgery associated with better – food intake after 60d (P ¼ .02) 16,536 11,720 (P ¼ .049)

a

Covered versus uncovered SEMS not specified. b Major complications after stent placement included stent migration (1) and obstruction (5). No major complications after surgery occurred. c Minor complications included post-procedure infection, delayed gastric emptying, and ileus. Abbreviations: pts, patients; OGJJ, open gastrojejunostomy; LGJJ, laparoscopic gastrojejunostomy; SEMS, self-expanding metal stent. All available statistically significant P values shown.

A. Stark and O.J. Hines

Endoscopic and operative palliation strategies

171

Similarly, a large retrospective series by Gray et al identified the presence of distant metastasis, poor tumor differentiation, severe preoperative nausea and vomiting, and lack of preoperative biliary stent as being significant risk factors for poor survival (o6 months) after palliative surgical bypass. The authors concluded that patients with 1 or more of the above risk factors should be considered strongly for nonoperative palliation.58 A comprehensive algorithm based on the above data is presented in Figure 2.

IS THERE A ROLE FOR PROPHYLACTIC SURGICAL BYPASS IN PATIENTS FOUND TO BE UNRESECTABLE AT TIME OF SURGERY? Despite advances in preoperative imaging, as many as 33% of patients with adenocarcinoma of the head of the pancreas will be found to be unresectable at the time of laparotomy, either due to unrecognized metastasis or locally advanced disease precluding resection.6 As obstructive jaundice

is commonly part of the presenting symptoms of the patient with unresectable disease, biliary bypass is often performed. Because 10%–-25% will go on to develop GOO, double loop bypass (DLB) frequently has been performed as a prophylactic operation when unresectable disease is found at the time of laparotomy. This approach is supporteed by data indicating a higher rate of reoperation in patients undergoing biliary bypass alone when compared to patients undergoing DLB.33,59 Two randomized trials have found benefit to performing prophylactic gastrojejunostomy at the time of laparotomy. The Hopkins group randomized those found to be unresectable at the time of laparotomy (excluding those with current or impending GOO) to biliary bypass or DLB. No differences in complications, length of stay, or survival were found. The cohort that underwent biliary bypass alone had a 19% rate of late GOO, versus 0% in the DLB group.39 The second study out of the Netherlands was halted after an interim analysis, partly in response to the results of the

PDAC with nausea/voming or preoperave evidence of gastric outlet obstrucon

Yes

Expected survival >2 months?

No

Potenally Resectable?

Surgical exploraon

No Sll resectable?

Clinical or laboratory evidence of biliary obstrucon?

Gastrojejunostomy

Consider laparoscopic gastrojejunostomy

Clinical or laboratory evidence of biliary obstrucon?

Resecon

No

No

Yes

Yes

No

Yes

Endoscopic uncovered duodenal stent (Consider prophylacc biliary stent)

Endoscopic biliary stent followed by endoscopic duodenal stent

Yes

Double loop bypass

Stent dysfuncon or failure?

Technical failure

Aempt repeat endoscopic management prior to surgery

Figure 2. Comprehensive algorithm for surgical and endoscopic management of malignant duodenal or gastric outlet obstruction related to pancreatic adenocarcinoma.

172

Hopkins group described above. They found that in unresectable patients without evidence of GOO, patients randomized to DLB experienced no increase in morbidity, mortality, length of stay, or survival when compared to biliary bypass alone. The number needed to be treated to prevent one reoperation for GOO was 6. Interestingly, there were no significant differences in quality of life between groups.41 The results of these studies were combined in a Cochrane review with similar overall conclusions.60 A contemporary, large VA database study also found similar rates of morbidity and mortality between patients undergoing DLB versus biliary bypass alone, however they found that reoperations for GOO were very low in patients with stage III and IV disease, secondary to short survival. The authors therefore recommended DLB in patients with stage III and IV disease only if symptomatic or found to have impending GOO at the time of operation.61 To challenge the above conclusions, a consecutive series of Memorial Sloan-Kettering patients from 1993–1997 found to have unresectable disease during staging were followed without any prophylactic intervention. Of the 151 patients, only three required surgery, two for GOO. Fourteen underwent endoscopic biliary drainage.62 These data suggest that a wait-and-see approach may be advisable. A thought provoking study from the Netherlands attempted to decrease morbidity and improve outcomes in this subset of patients. A policy of routine laparoscopic staging prior to laparotomy was implemented; those patients found to be unresectable via laparoscopy were then randomized to surgical DLB or endoscopic biliary stent placement. Overall, no differences were found between groups, but this may be a manifestation of the abysmal (35%) diagnostic yield of laparoscopy in predicting unresectability, rather than an equivalency of morbidity between prophylactic non-operative and operative palliation.63 Finally, a recent retrospective study by Lyons et al examined 157 patients after nontherapeutic laparotomy (21% laparotomy only, biliary bypass 30%, duodenal bypass 11%, DLB 38%). There were no differences in total in-hospital days between all operations, and no operation was predictive of decreased need for post-operative intervention. Furthermore, regardless of operation performed, preoperative endoscopic biliary drainage was associated with decreased post-operative complications.9

ENDOSCOPY FOR COMBINED BILIARY AND DUODENAL OBSTRUCTION Previously considered an indication for operative palliation, numerous studies have established the efficacy and safety of performing combined endoscopic biliary and duodenal stent placement for

A. Stark and O.J. Hines

simultaneous malignant biliary and gastric outlet obstruction. A series from the MD Anderson reported that 17/18 patients successfully underwent combined endoscopic biliary and duodenal stent placement. This was accomplished despite advanced disease, as balloon dilation was required in eight to allow passage of the endoscope. One required surgical DLB secondary to a tortuous stricture of the duodenum. Thirteen patients tolerated a regular diet thereafter.64 Concerns regarding the ability to access the common bile duct after duodenal stenting have been slowly put to rest as endoscopists gain more experience with these challenging cases. In a retrospective study of 100 unresectable patients, 23 developed both biliary and duodenal obstruction, 21 had successful combined endoscopic relief including the patients with early duodenal obstruction.4

PALLIATIVE RESECTION HAS NO ROLE IN THE AGE OF ENDOSCOPIC PALLIATION In 1996, Lillemoe et al published results of 64 patients that underwent pancreaticoduodenectomy with grossly or microscopically positive margins, and compared the outcomes with a matched cohort of surgically palliated patients with unresectable disease. They were able to show that palliative resection was not associated with higher complications, postoperative mortality, or differences in adjuvant chemotherapy. Operative mortality was 1.2%. There was a significant trend towards improved survival in those patients that underwent palliative resection versus surgical palliation (15.3 months v 11.2 months, P o.02).65 This survival benefit, though small, served as a rationale to promote the role of palliative Whipple in patients with incurable disease. These results have been subsequently challenged by a more recent retrospective series comparing palliative resection to DLB. Schniewind et al separated the group undergoing DLB into those with metastatic disease, and those with locally advanced disease. No significant overall difference in mortality was seen between palliative resection and DLB; a small, statistically insignificant difference in mean survival between resection and palliation was completely obviated when comparing the resection group to the locally advanced cohort of DLB patients. This may indicate that the survival advantage found was spurious. No differences in quality of life were found. Morbidity and mortality were 58% and 7.8%, respectively, calling into question the reproducibility of the results of the Hopkins trial described above.66 For these reasons, palliative resection is not recommended as a therapy for palliation of advanced disease, especially in light of the availability of less invasive and less morbid palliative options.

Endoscopic and operative palliation strategies

SURGICAL AND ENDOSCOPIC MANAGEMENT OF TUMOR-ASSOCIATED PAIN Pain occurs in a majority of patients with pancreatic adenocarcinoma, and is treated with opioid analgesics. Severe, intractable pain is often related to malignant invasion of the celiac plexus. Accordingly, celiac plexus block has been a common component of operative and non-operative palliation strategies. Side effects include urinary retention, back pain, diarrhea, and hypotension. Serious complications such as transient or permanent paraplegia are extremely rare. Intraoperative, percutaneous, or endoscopic ultrasound (EUS)-guided celiac plexus block may be performed. When unresectable disease is encountered during operation, the technique for intraoperative celiac neurolysis involves injection of either diluted ethanol or anesthetic (bupivacaine or lidocaine) to either side of the aorta at the level of the celiac axis. A highly influential, double-blinded, randomized trial by Lillemoe et al studied the efficacy of intraoperative celiac neurolysis with 50% ethanol. One hundred thirty-five patients were intraoperatively randomized once determined to be unresectable; half of the patients received celiac neurolysis with injection of 50% ethanol, and half received placebo normal saline injection. Overall, the treatment group —regardless of the presence of pain preoperatively —had a statistically significant decrease in pain compared to placebo at 2, 4, and 6 months after operation. Most interestingly, although overall survival was similar between groups, in the subset of patients with preoperative pain there was a significant survival advantage.67 A later, non-randomized comparison of intraoperative celiac plexus block with 1% lidocaine with pharmacological therapy alone found a significant and durable decrease in overall opioid consumption in patients undergoing operative celiac plexus block.68 Division of the greater splanchnic nerve via a thoracoscopic approach also has been advocated as a durable option for treating tumor associated pain. Small series demonstrated duration of pain relief at 1 and 3 months postoperatively, with complete relief occurring in up to 84% of patients. Complications were uncommon, perhaps secondary to the small sample size.69,70 EUS guided celiac plexus block has been demonstrated to be effective. The only randomized trial of EUS versus percutaneous celiac plexus block was performed in patients with chronic pancreatitis. EUS was demonstrated to be safe and effective, with perhaps an even more persistent relief of pain than the percutaneous approach.71 A meta-analysis of eight studies examining EUS celiac plexus block for patients with pancreatic cancer found that there was

173

an overall 85% response to therapy with very low rates of complication.72 Celiac plexus block versus analgesic therapy alone was the subject of a recent Cochrane review. Inclusive of six RCTs, the meta-analysis found that celiac plexus block had a small, but significant decrease in subjective pain at 4 and 8 weeks postprocedure. However, the most compelling result was a durable decrease in overall opioid consumption, which translated into a significantly decreased rate of opioid related side effects, notably constipation.73 The criticism of many of these trials—including Lillemoe et al—is that post-procedure analgesic therapy was not standardized but rather left to individual physician discretion. A double-blinded, randomized trial addressed this concern specifically by randomizing patients to optimized systemic analgesic therapy with or without celiac plexus block. Oral analgesics were administered according to an algorithm. Although both groups responded favorably to treatment, they found that fewer patients reported moderate or severe pain during the first 6 weeks after celiac plexus block. No significant differences in opioid consumption, opioid related side effects, quality of life, or survival were found between groups.74

CONCLUSION AND RECOMMENDATIONS Patients with unresectable adenocarcinoma of the head of the pancreas frequently require intervention for palliation of biliary obstruction, duodenal or gastric outlet obstruction, and tumor-associated pain. In these patients with short expected survival, interventions ideally will provide durable relief that maximizes quality of life while minimizing morbidity and future interventions. In this review, we have examined the historical and current evidence supporting both operative and non-operative palliative interventions. Open surgery was once the gold standard, and palliative biliary and gastric bypass often was performed prophylactically; the paradigm has now shifted towards less invasive and endoscopic modes of palliation. This is in part a result of the development of self-expanding metal stents that provide for effective palliation of biliary obstruction with significantly reduced need for re-intervention when compared to earlier plastic stents. It also in part is due to the growing experience of endoscopists leading to high rates of technical success in placing both biliary and duodenal stents with minimal morbidity. Thus in centers with the necessary expertise, an initial non-operative approach to palliation should be attempted in the preoperatively identified unresectable patient with pancreatic adenocarcinoma. This approach allows for these conditions to be addressed quickly and for the

174

patient to initiate or get back to treatment with chemotherapy and/or radiation. Surgery remains an effective method of palliation, and should be performed in patients with disease that is not amenable to endoscopy.

A. Stark and O.J. Hines

15.

16.

REFERENCES 1. Siegel R, Miller KD, Jemal A. Cancer statistics, 2015. CA. 2015;65:5–29. 2. Howlader N, Noone A, Krapcho M, et al (eds). SEER Cancer Statistics Review, 1975-2010, National Cancer Institute. Bethesda, MD, http://seer.cancer.gov/csr/ 1975_2010/, based on November 2012 SEER data submission, posted to the SEER web site, April 2013. 3. Schneider G, Siveke JT, Eckel F, Schmid RM. Pancreatic cancer: basic and clinical aspects. Gastroenterology. 2005;128:1606–25. 4. Maire F, Hammel P, Ponsot P, et al. Long-term outcome of biliary and duodenal stents in palliative treatment of patients with unresectable adenocarcinoma of the head of pancreas. Am J Gastroenterol. 2006;101: 735–42. 5. Kneuertz PJ, Cunningham SC, Cameron JL, et al. Palliative surgical management of patients with unresectable pancreatic adenocarcinoma: trends and lessons learned from a large, single institution experience. J Gastrointest Surg. 2011;15:1917–27. 6. Sohn TA, Lillemoe KD, Cameron JL, Huang JJ, Pitt HA, Yeo CJ. Surgical palliation of unresectable periampullary adenocarcinoma in the 1990s. J Am Coll Surg. 1999;188:658–66. 7. Singh SM, Longmire WP, Jr., Reber HA. Surgical palliation for pancreatic cancer. The UCLA experience. Ann Surg. 1990;212:132–9. 8. Sarr MG, Cameron JL. Surgical management of unresectable carcinoma of the pancreas. Surgery. 1982;91:123–33. 9. Lyons JM, Karkar A, Correa-Gallego CC, et al. Operative procedures for unresectable pancreatic cancer: does operative bypass decrease requirements for postoperative procedures and in-hospital days? HPB. 2012;14:469–75. 10. Spanheimer PM, Cyr AR, Liao J, et al. Complications and survival associated with operative procedures in patients with unresectable pancreatic head adenocarcinoma. J Surg Oncol. 2014;109:697-701. 11. Huguier M, Baumel H, Manderscheid JC, Houry S, Fabre JM. Surgical palliation for unresected cancer of the exocrine pancreas. Eur J Surg Oncol. 1993;19: 342–7. 12. Sarfeh IJ, Rypins EB, Jakowatz JG, Juler GL. A prospective, randomized clinical investigation of cholecystoenterostomy and choledochoenterostomy. Am J Surg. 1988;155:411–4. 13. Rosemurgy AS, Burnett CM, Wasselle JA. A comparison of choledochoenteric bypass and cholecystoenteric bypass in patients with biliary obstruction due to pancreatic cancer. Am Surg. 1989;55:55–60. 14. Dumonceau JM, Tringali A, Blero D, et al. Biliary stenting: indications, choice of stents and results: European Society of Gastrointestinal Endoscopy

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

(ESGE) clinical guideline. Endoscopy. 2012;44: 277–98. Andersen JR, Sorensen SM, Kruse A, Rokkjaer M, Matzen P. Randomised trial of endoscopic endoprosthesis versus operative bypass in malignant obstructive jaundice. Gut. 1989;30:1132–5. Kahaleh M, Talreja JP, Loren DE, et al. Evaluation of a fully covered self-expanding metal stent with flared ends in malignant biliary obstruction: a multicenter study. J Clin Gastroenterol. 2013;47:e96–100. Soderlund C, Linder S. Covered metal versus plastic stents for malignant common bile duct stenosis: a prospective, randomized, controlled trial. Gastrointest Endosc. 2006;63:986–95. 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. Shepherd HA, Royle G, Ross AP, Diba A, Arthur M, ColinJones D. Endoscopic biliary endoprosthesis in the palliation of malignant obstruction of the distal common bile duct: a randomized trial. Br J Surg. 1988;75:1166–8. Sung JJ, Sollano JD, Lai CW, et al. Long-term ciprofloxacin treatment for the prevention of biliary stent blockage: a prospective randomized study. Am J Gastroenterol. 1999;94:3197–201. Kitano M, Yamashita Y, Tanaka K, et al. Covered selfexpandable metal stents with an anti-migration system improve patency duration without increased complications compared with uncovered stents for distal biliary obstruction caused by pancreatic carcinoma: a randomized multicenter trial. Am J Gastroenterol. 2013;108:1713–22. Moss AC ME, MacMathuna P. Palliative biliary stents for obstructing pancreatic carcinoma. Cochrane Database of Systematic Reviews. 2006. Knyrim K, Wagner HJ, Pausch J, Vakil N. A prospective, randomized, controlled trial of metal stents for malignant obstruction of the common bile duct. Endoscopy. 1993;25:207–12. Adams MA, Anderson MA, Myles JD, Khalatbari S, Scheiman JM. Self-expanding metal stents (SEMS) provide superior outcomes compared to plastic stents for pancreatic cancer patients undergoing neoadjuvant therapy. J Gastrointest Incol. 2012;3:309–13. Davids PH, Groen AK, Rauws EA, Tytgat GN, Huibregtse K. Randomised trial of self-expanding metal stents versus polyethylene stents for distal malignant biliary obstruction. Lancet. 1992;340:1488–92. Cho JH, Jeon TJ, Park JY, et al. Comparison of outcomes among secondary covered metallic, uncovered metallic, and plastic biliary stents in treating occluded primary metallic stents in malignant distal biliary obstruction. Surg Endosc. 2011;25:475–82. Hamada T, Isayama H, Nakai Y, et al. Duodenal invasion is a risk factor for the early dysfunction of biliary metal stents in unresectable pancreatic cancer. Gastrointest Endosc. 2011;74:548–55. van der Gaag NA, Rauws EA, van Eijck CH, et al. Preoperative biliary drainage for cancer of the head of the pancreas. N Engl J Med. 2010;362:129–37.

Endoscopic and operative palliation strategies

29. Brandabur JJ, Kozarek RA, Ball TJ, et al. Nonoperative versus operative treatment of obstructive jaundice in pancreatic cancer: cost and survival analysis. Am J Gastroenterol. 1988;83:1132–9. 30. Scott EN, Garcea G, Doucas H, Steward WP, Dennison AR, Berry DP. Surgical bypass v endoscopic stenting for pancreatic ductal adenocarcinoma. HPB. 2009;11:118–24. 31. Raikar GV, Melin MM, Ress A, et al. Cost-effective analysis of surgical palliation versus endoscopic stenting in the management of unresectable pancreatic cancer. Ann Surg Oncol. 1996;3:470–5. 32. 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;344:1655–60. 33. Schwarz A, Beger HG. Biliary and gastric bypass or stenting in nonresectable periampullary cancer: analysis on the basis of controlled trials. Int J Pancreatol. 2000;27:51–8. 34. Castano R, Lopes TL, Alvarez O, Calvo V, Luz LP, Artifon EL. Nitinol biliary stent versus surgery for palliation of distal malignant biliary obstruction. Surg Endosc. 2010;24:2092–8. 35. Artifon EL, Sakai P, Cunha JE, et al. Surgery or endoscopy for palliation of biliary obstruction due to metastatic pancreatic cancer. Am J Gastroenterol. 2006;101:2031–7. 36. Moss AC, Morris E, Leyden J, MacMathuna P. Malignant distal biliary obstruction: a systematic review and meta-analysis of endoscopic and surgical bypass results. Cancer Treat Rev. 2007;33:213–21. 37. Glazer ES, Hornbrook MC, Krouse RS. A meta-analysis of randomized trials: immediate stent placement v surgical bypass in the palliative management of malignant biliary obstruction. J Pain Symptom Manag. 2014; 47:307–14. 38. Weaver DW, Wiencek RG, Bouwman DL, Walt AJ. Gastrojejunostomy: is it helpful for patients with pancreatic cancer? Surgery. 1987;102:608–13. 39. Lillemoe KD, Cameron JL, Hardacre JM, et al. Is prophylactic gastrojejunostomy indicated for unresectable periampullary cancer? A prospective randomized trial Ann Surg. 1999;230:322–8. 40. House MG, Choti MA. Palliative therapy for pancreatic/ biliary cancer. Surg Clin North Am. 2005;85:359–71. 41. Van Heek NT, De Castro SM, van Eijck CH, 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;238:894–902. 42. Kazanjian KK, Reber HA, Hines OJ. Laparoscopic gastrojejunostomy for gastric outlet obstruction in pancreatic cancer. Am Surg. 2004;70:910–3. 43. Alam TA, Baines M, Parker MC. The management of gastric outlet obstruction secondary to inoperable cancer. Surg Endosc. 2003;17:320–3. 44. Navarra G, Musolino C, Venneri A, De Marco ML, Bartolotta M. Palliative antecolic isoperistaltic gastrojejunostomy: a randomized controlled trial comparing open and laparoscopic approaches. Surg Endosc. 2006;20:1831–4.

175

45. Espinel J, Vivas S, Munoz F, Jorquera F, Olcoz JL. Palliative treatment of malignant obstruction of gastric outlet using an endoscopically placed enteral Wallstent. Dig Dis Sci. 2001;46:2322–4. 46. Woo SM, Kim DH, Lee WJ, et al. Comparison of uncovered and covered stents for the treatment of malignant duodenal obstruction caused by pancreaticobiliary cancer. Surg Endosc. 2013;27:2031–9. 47. Kaw M, Singh S, Gagneja H, Azad P. Role of selfexpandable metal stents in the palliation of malignant duodenal obstruction. Surg Endosc. 2003;17:646–50. 48. Dormann A, Meisner S, Verin N, Wenk Lang A. Selfexpanding metal stents for gastroduodenal malignancies: systematic review of their clinical effectiveness. Endoscopy. 2004;36:543–50. 49. Graber I, Dumas R, Filoche B, et al. The efficacy and safety of duodenal stenting: a prospective multicenter study. Endoscopy. 2007;39:784–7. 50. Wong YT, Brams DM, Munson L, et al. Gastric outlet obstruction secondary to pancreatic cancer: surgical v endoscopic palliation. Surg Endosc. 2002;16:310–2. 51. Fiori E, Lamazza A, Volpino P, et al. Palliative management of malignant antro-pyloric strictures. Gastroenterostomy v endoscopic stenting. A randomized prospective trial. Anticancer Res. 2004;24:269–71. 52. Jeurnink SM, van Eijck CH, Steyerberg EW, Kuipers EJ, Siersema PD. Stent versus gastrojejunostomy for the palliation of gastric outlet obstruction: a systematic review. BMC Gastroenterol. 2007;7:18. 53. Ly J, O'Grady G, Mittal A, Plank L, Windsor JA. A systematic review of methods to palliate malignant gastric outlet obstruction. Surg Endosc. 2010;24:290–7. 54. Chandrasegaram MD, Eslick GD, Mansfield CO, et al. Endoscopic stenting versus operative gastrojejunostomy for malignant gastric outlet obstruction. Surg Endosc. 2012;26:323–9. 55. Jeurnink SM, Steyerberg EW, van Hooft JE, et al. Surgical gastrojejunostomy or endoscopic stent placement for the palliation of malignant gastric outlet obstruction (SUSTENT study): a multicenter randomized trial. Gastrointest Endosc. 2010;71:490–9. 56. Mehta S, Hindmarsh A, Cheong E, et al. Prospective randomized trial of laparoscopic gastrojejunostomy versus duodenal stenting for malignant gastric outflow obstruction. Surg Endosc. 2006;20:239–42. 57. Jeurnink SM, Steyerberg EW, Vleggaar FP, et al. Predictors of survival in patients with malignant gastric outlet obstruction: a patient-oriented decision approach for palliative treatment. Dig Liver Dis. 2011;43:548–52. 58. Gray PJ, Jr, Wang J, Pawlik TM, et al. Factors influencing survival in patients undergoing palliative bypass for pancreatic adenocarcinoma. J Surg Oncol. 2012;106:66–71. 59. Neuberger TJ, Wade TP, Swope TJ, Virgo KS, Johnson FE. Palliative operations for pancreatic cancer in the hospitals of the U.S. Department of Veterans Affairs from 1987 to 1991. Am J Surg. 1993;166:632–6. 60. Gurusamy KS, Kumar S, Davidson BR. Prophylactic gastrojejunostomy for unresectable periampullary carcinoma. Cochrane Database Systematic Rev. 2013;2: CD008533.

176

61. Wade TP, Neuberger TJ, Swope TJ, Virgo KS, Johnson FE. Pancreatic cancer palliation: using tumor stage to select appropriate operation. Am J Surg. 1994;167: 208–12. 62. Espat NJ, Brennan MF, Conlon KC. Patients with laparoscopically staged unresectable pancreatic adenocarcinoma do not require subsequent surgical biliary or gastric bypass. J Am Coll Surg. 1999;188: 649–55. 63. Nieveen van Dijkum EJ, Romijn MG, Terwee CB, et al. Laparoscopic staging and subsequent palliation in patients with peripancreatic carcinoma. Ann Surg. 2003;237:66–73. 64. Kaw M, Singh S, Gagneja H. Clinical outcome of simultaneous self-expandable metal stents for palliation of malignant biliary and duodenal obstruction. Surg Endosc. 2003;17:457–61. 65. Lillemoe KD, Cameron JL, Yeo CJ, et al. Pancreaticoduodenectomy. Does it have a role in the palliation of pancreatic cancer? Ann Surg. 1996;223:718–25. 66. Schniewind B, Bestmann B, Kurdow R, et al. Bypass surgery versus palliative pancreaticoduodenectomy in patients with advanced ductal adenocarcinoma of the pancreatic head, with an emphasis on quality of life analyses. Ann Surg Oncol. 2006;13:1403–11. 67. 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;217:447–55. 68. Okuyama M, Shibata T, Morita T, et al. A comparison of intraoperative celiac plexus block with

A. Stark and O.J. Hines

69.

70.

71.

72.

73.

74.

pharmacological therapy as a treatment for pain of unresectable pancreatic cancer. J Hepatobiliary Pancreatic Surg. 2002;9:372–5. Pietrabissa A, Vistoli F, Carobbi A, Boggi U, Bisa M, Mosca F. Thoracoscopic splanchnicectomy for pain relief in unresectable pancreatic cancer. Arch Surg. 2000;135:332–5. Saenz A, Kuriansky J, Salvador L, et al. Thoracoscopic splanchnicectomy for pain control in patients with unresectable carcinoma of the pancreas. Surg Endosc. 2000;14:717–20. 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;94:900–5. Puli SR, Reddy JB, Bechtold ML, Antillon MR, Brugge WR. EUS-guided celiac plexus neurolysis for pain due to chronic pancreatitis or pancreatic cancer pain: a meta-analysis and systematic review. Dig Dis Sci. 2009;54:2330–7. Arcidiacono PG, Calori G, Carrara S, McNicol ED, Testoni PA. Celiac plexus block for pancreatic cancer pain in adults. Cochrane Database Systematic Rev. 2011:CD007519. Wong GY, Schroeder DR, Carns PE, et al. Effect of neurolytic celiac plexus block on pain relief, quality of life, and survival in patients with unresectable pancreatic cancer: a randomized controlled trial. JAMA. 2004;291:1092–9.