| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIALS |
From the Hepatobiliary Unit, Department of Surgery, Manchester Royal Infirmary, Manchester, United Kingdom.
Correspondence: Address correspondence to: A. K. Siriwardena, MD, HPB Unit, Department of Surgery, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, United Kingdom; Fax: 0161-276-4530; E-mail: ajith.siriwardena{at}cmmc.nhs.uk
An estimated 30,300 new cases of pancreatic cancer diagnosed in the United States in 2002; of these, 29,700 deaths occurred.1 As the mortality rate closely approximates the incidence, survival is poor.1 In his recent editorial, Lowy2 concluded that despite advances in the understanding of molecular heterogeneity and gene expression in pancreatic cancer, the prognosis remains poor. Currently, surgical resection is the only therapeutic option associated with prolonged survival.3 Despite refinements in imaging and staging, most patients with pancreatic cancer are unsuitable for resection, either because of comorbidity or the presence of locally advanced tumor or metastatic disease.4
When curative treatment is unfeasible, careful selection of optimal palliation becomes of central importance in the management of pancreatic cancer. Further, in contemporary management, optimization of quality of life in patients with nonresectable disease is an important goal. Thus, the rational focus of care in patients with nonresectable disease is on palliation of symptoms using therapeutic interventions associated with the least morbidity.
Trends in management over the last decade have favored a shift toward endoscopic palliation5,6 as these techniques avoid surgery and do not require prolonged hospitalization. Proponents of surgical bypass would argue that a role remains for operative biliary-enteric anastomosis in younger patients with inoperable cancer and low comorbidity because these individuals are likely to have survival times sufficient to lead to problems with multiple episodes of stent occlusion were they to be treated endoscopically. A typical management algorithm based on these approaches is seen in Figure 1.7
|
In its simplest (and probably most widely applied form), laparoscopic biliary bypass takes the form of an intracorporeally stapled anastomosis between the gallbladder and a loop of proximal small bowel (i.e., cholecystojejunostomy).9–11 Advocates of laparoscopic intervention state that laparoscopic cholecystojejunostomy can be incorporated at the time of staging if evidence of irresectability exists and that the minimal intervention approach of this type of procedure realigns the management algorithm toward surgery.
Paradoxically, in the era of open surgery, a general shift of opinion had been away from cholecystojejunostomy toward bypass to the bile duct because biliary-enteric patency rates were thought to be improved by avoiding dependence on cystic duct patency. Support for this trend came from a small randomized comparison of open cholecystoenterostomy with bile duct bypass, which revealed a higher incidence of recurrent jaundice in patients having bypass to the gallbladder.12 Although no adequately powered comparative trials exist, the recent study by Urbach et al.13 provides novel and important information on this issue. These authors undertook a retrospective cohort study based on North American Medicare Claims and Surveillance, Epidemiology and End Results (SEER) data. Comparing patients with nonresectable pancreatic cancer having intestinal bypass to the gallbladder with those having bypass to the bile duct, they found a 4.4 times increase in the risk of requiring subsequent biliary drainage procedures in those individuals having bypass to the gallbladder. In addition, median survival was longer in patients having bypass to the bile duct, with an adjusted hazard ratio for death associated with bypass to the gallbladder of 1.2 (possibly related to problems of sepsis from inadequate biliary drainage). It should be noted that the study does not specify whether the laparoscopic or open route was selected; however, given that the data were collected for the period from 1991 to 1996, it is likely that the patients had open surgery. As the nature of biliary drainage achieved at open cholecystojejunostomy is fundamentally similar to that achieved laparoscopically, extrapolation is valid.
Probably the most frequent cause for a functioning cholecystoenterostomy failing to provide sustained relief of jaundice is occlusion of the cystic duct by tumor progression. In this context, Tarnasky et al.14 carried out a retrospective evaluation of endoscopic retrograde cholangiopancreatography (ERCP) findings in patients with malignant obstructive jaundice: of 218 patients having the procedure, 102 were excluded as they had either hilar strictures or prior biliary surgery. Further exclusions left 50 patients with radiologically patent hepatocystic junctions of whom 22 were >1 cm from the upper limit of the obstruction. These authors conclude firstly that only a minority (22 of 218) of patients with malignant nonhilar obstructive jaundice are suitable for bypass to the gallbladder. Of this subgroup of 22 with patent hepatocystic junctions and a distance of >1 cm from junction to obstruction, 4 had ampullary tumors and 5 had distal bile duct cholangiocarcinomas and, thus, in all probability would have been candidates for surgical resection. Importantly, they also provide evidence that proof of a patent hepatocystic junction should be established before bypass.
Integrating these studies into modern management algorithms for patients with malignant obstructive, laparoscopic biliary bypass appears to fail two simple tests of good medical practice. First, the test for application of a novel laparoscopic procedure: does the proposed operation duplicate the standard open procedure modifying only the route of access? In this case, laparoscopic cholecystojejunostomy appears to provide inferior palliation of jaundice compared with bypass to the bile duct—probably the "standard" surgical bypass in current practice.15
The second test is that of avoiding unnecessary intervention in patients with incurable cancer. In this regard, palliation of jaundice seems well served by endobiliary stenting.16 Further, the cost-effectiveness advantages demonstrated by endoscopic stenting in comparison with open surgical bypass are likely to translate to the laparoscopic biliary bypass because a principal component of cost relates to the need for in-patient care.17 Further, ongoing advances in biotechnology suggest that metallic stents can reduce the frequency of stent occlusion compared with plastic stents18 and those few patients who develop duodenal obstruction as a nonagonal event can be treated by duodenal stenting,19,20 although here the laparoscopic route may be logical for gastric bypass.21
Laparoscopic Roux-en-Y hepaticojejunostomy22 has the theoretic attractions of avoiding the disadvantages of cholecystojejunostomy while retaining the advantages of the laparoscopic approach, in particular, the option to combine staging with treatment in patients found at laparoscopy to have unresectable disease. In the Röthlin et al.22 series of 14 patients having laparoscopic hepaticojejunostomy, the median operating time was 129 minutes and no perioperative deaths occurred. The median in-patient stay, however, was 9 days and outcome was compared with a historic cohort treated by open surgery rather than patients treated endoscopically. Further knowledge of the feasibility and general applicability of this technique and outcome in comparison with current best endoscopic care is required before this procedure can generally be recommended.
In summary, cancer care in the 21st century should utilize available evidence optimally. The thirst for technologic advance in minimally invasive surgery should not result in a return to cholecystojejunostomy—a procedure that is applicable to only a minority of patients with pancreatic cancer and which provides inferior relief of jaundice.
Received for publication December 5, 2003. Accepted for publication July 5, 2004.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
