Measures to Prevent Pancreatic Fistula After Pancreatoduodenectomy: A Comprehensive Review | Gastroenterology | JAMA Surgery | JAMA Network
[Skip to Navigation]
Sign In
Table 1. 
Randomized Controlled Trials Comparing Outcomes of the Use of Somatostatin and Its Analogues in Pancreatic Resection
Randomized Controlled Trials Comparing Outcomes of the Use of Somatostatin and Its Analogues in Pancreatic Resection
Table 2. 
Randomized Controlled Trials Comparing Pancreaticogastrostomy vs Pancreaticojejunostomy
Randomized Controlled Trials Comparing Pancreaticogastrostomy vs Pancreaticojejunostomy
November 16, 2009

Measures to Prevent Pancreatic Fistula After Pancreatoduodenectomy: A Comprehensive Review

Author Affiliations

Author Affiliations: Faculty of Medicine, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, China.

Arch Surg. 2009;144(11):1074-1080. doi:10.1001/archsurg.2009.193

Objective  To review the current evidence on the efficacy of various interventions directed toward prevention of pancreatic fistula after pancreatoduodenectomy.

Data Sources  A MEDLINE database search from January 1, 1990, to August 31, 2008, was performed to identify relevant articles using the keywords pancreatoduodenectomy, pancreaticojejunostomy, pancreaticogastrostomy, pancreatic anastomotic leakage, and pancreatic fistula. Additional articles were identified by a manual search of the references from the key articles.

Study Selection  Case reports were excluded from this study.

Data Extraction  Of the identified studies, only those published in English describing meta-analyses or randomized controlled trials were considered first. In those aspects with limited or no randomized controlled trials, nonrandomized comparative studies and case series were included also.

Data Synthesis  The prophylactic use of perioperative somatostatin and its analogues to prevent pancreas-related complications after pancreatic surgery remains controversial and does not result in a reduction of operative mortality. There is no clear evidence for or against a particular type of pancreaticoenteric anastomosis. Pancreaticoenterostomy cannot be replaced by pancreatic ductal obliteration. Pancreaticogastrostomy is equivalent to pancreaticojejunostomy in perioperative morbidity and mortality.

Conclusion  More large-scale comparative studies and randomized controlled trials are required to determine the optimum pharmacologic interventions and technique of pancreaticoenteric anastomosis after pancreatoduodenectomy.

In 1898, Halsted performed the first local excision of carcinoma of ampulla of Vater.1 In 1909, the first successful regional resection of a periampullary tumor was performed by Kausch.2 He performed the operation as a 2-stage procedure in which a cholecystojejunostomy was performed 6 weeks before the second operation, which involved resection of the head of the pancreas, the pylorus, and the first and second parts of the duodenum, followed by gastroenterostomy, closure of the distal common bile duct, and anastomosis of the residual pancreas to the third part of the duodenum. Resection of periampullary tumor was popularized in a 1935 article by Whipple and colleagues.3 Their 2-stage pancreatoduodenectomy consisted of posterior gastroenterostomy, ligation and division of the common bile duct and cholecystogastrostomy in the first stage, followed by resection of the duodenum and pancreatic head in the second stage. The pancreatic stump was closed with sutures, without a pancreaticoenteric anastomosis. Whipple4 later completed the whole procedure in a single stage in 1940, and the reconstruction was modified in 1942 to include pancreaticojejunostomy, as he found a high rate of pancreatic fistula after closure of pancreatic stumps.5

Improvements in anesthesia and perioperative management decreased the surgical mortality to less than 5% for pancreatoduodenectomy.6-12 Unfortunately, the postoperative morbidity remains high and approaches 50% even in large series.6-12 The key determinant of outcome after pancreatoduodenectomy remains the pancreaticoenteric anastomosis. Sepsis and hemorrhage, caused by pancreatic fistula, are associated with mortality of 20% to 40%6-12 and result in prolonged hospitalization and increased hospital expenses. The terminology in the medical literature to describe pancreatic fistula, pancreatic leakage, or anastomotic insufficiency is confusing, and the terms have been interchangeably used. The reported pancreatic fistula rate ranged from 2% to 22% depending on the definition used.6-12 However, there is no universally accepted definition of pancreatic fistula. A uniform definition is urgently needed to make results comparable between different series. An international panel of pancreatic surgeons from well-known centers (the International Study Group on Pancreatic Fistula Definition12) worked together to develop a simple, objective, and easy-to-apply definition of postoperative pancreatic fistula, graded primarily on clinical effect. An all-inclusive definition is a drain output of any measurable volume of fluid on or after postoperative day 3 with an amylase content greater than 3 times the serum amylase activity.

Several pharmacologic and technical interventions have been suggested to decrease the pancreatic fistula rate, but the results have been controversial. The only consistently reproducible factor is the establishment of specialist centers performing pancreatic surgery in high volume.13-18 This article reviews the current evidence on the efficacy of various interventions directed toward prevention of pancreatic fistula after pancreatoduodenectomy.

Search strategy

A MEDLINE database search from January 1, 1990, to August 31, 2008, was performed to identify relevant articles using the keywords pancreatoduodenectomy, pancreaticojejunostomy, pancreaticogastrostomy, pancreatic anastomotic leakage, and pancreatic fistula. Additional articles were identified by a manual search of the references from the key articles.

Of the identified studies, only those published in English describing meta-analyses or randomized controlled trials (RCTs) were considered first. In those aspects with limited or no RCTs, nonrandomized comparative studies and case series were included also. Only full peer-reviewed articles were selected. Two of us (E.C.H.L. and W.Y.L.) independently identified and screened the search findings for potentially eligible articles. Any disagreement was discussed and a consensus reached.

Inclusion and exclusion criteria

Only studies about pancreatoduodenectomy reporting the outcomes of pancreatic fistula rates or pancreatic-related complications were selected. Case reports were excluded.

Data analysis

Level of evidence was graded according to the following hierarchy of evidence developed by the National Health and Medical Research Council of Australia19 (hereafter evidence levels 1-4): (1) Evidence obtained from a systematic review of all relevant RCTs. (2) Evidence obtained from at least 1 properly designed RCT. (3) (a) Evidence obtained from well-designed pseudo-RCTs (alternate allocation or some other method). (b) Evidence obtained from cohort studies, case-control studies, interrupted time series with a control group, or comparative studies (including systematic reviews of such studies) with concurrent controls and allocation not randomized. (c) Evidence obtained from comparative studies with historical control, 2 or more single-arm studies, or interrupted time series without a parallel control group. (4) Evidence obtained from case series (either posttest or pretest and posttest).

Risk factors

In an attempt to prevent pancreatic fistula after pancreatoduodenectomy, some risk factors have been identified. These include pancreas risk factors (pancreatic texture, pancreatic duct size, blood supply to the stump, volume of pancreatic juice output, and pathologic features), patient risk factors (age, sex, bilirubin level, and comorbid illness), and operation risk factors (operating time, blood loss, type of anastomosis, and stent use).20-33 Among these risk factors, pancreatic texture, pancreatic duct size, blood supply to the stump, and volume of pancreatic juice output have been considered the most significant. It is generally accepted that a fibrotic pancreatic remnant facilitates the pancreaticoenteric anastomosis, whereas a soft pancreatic remnant frequently results in a higher pancreatic fistula rate.

Pharmacologic intervention

Somatostatin is a potent inhibitor of pancreatic endocrine and exocrine functions. The synthetic peptide octreotide contains the same amino acid sequences essential to the activity of somatostatin, while conferring resistance to enzyme degradation, resulting in a long- acting stable analog suitable for subcutaneous administration. The rationale is that, by decreasing the volume of pancreatic secretion, the pancreatic fistula rate would be decreased because the pancreaticoenteric anastomosis would heal better. Adverse effects of somatostatin and its analogues include nausea, flatulence, diarrhea, steatorrhea, pain at the injection site, and abdominal discomfort.

There were 11 RCTs involving 2023 patients that examined the use of somatostatin and its analogues (Table 1).34-44 Five RCTs from Europe34-37,40 and 1 RCT from Asia42 showed the benefit of perioperative use of somatostatin and its analogues to decrease the postoperative complication rate. On the other hand, 2 recent RCTs from Europe43,44 and 3 RCTs from the United States38,39,41 failed to show a benefit. Two recent meta-analyses45,46 have been published. Connor et al45 analyzed 10 studies and showed that somatostatin and its analogues did not reduce mortality after pancreatic surgery but reduced total morbidity and pancreas-specific complications. Somatostatin and its analogues reduced the rate of biochemical fistula but not the incidence of clinical anastomotic disruption. Alghamdi et al46 analyzed 7 studies and showed that perioperative octreotide administration was associated with a significant reduction of the pancreatic fistula rate after pancreatic surgery. However, this risk reduction was not associated with a significant difference in postoperative mortality. Conclusions drawn from these meta-analyses should be cautionary, as pooling of data from these RCTs was difficult because there was considerable heterogeneity in these studies for end point measures, definition of outcome measurements, treatment regimens, pathologic findings, types of pancreatic surgery, and anastomotic technique.

Prophylactic use of perioperative somatostatin and its analogues to prevent pancreas-related complications after pancreatic surgery remains controversial. It does not result in a reduction of operative mortality. There is a need for RCTs with standardization in definition of outcome measurements, treatment regimen, surgical technique, and stratification of risk factors (evidence levels 1 and 2).

Technical intervention

To prevent complications following pancreatoduodenectomy, various techniques of managing the pancreatic remnant have been proposed. These range from pancreatic ductal occlusion to pancreaticoenterostomy with the jejunum or stomach.

Pancreatic ductal occlusion

In an attempt to obviate a pancreaticoenteric anastomosis, pancreatic ductal occlusion was studied. There were 2 RCTs.47,48 In the study by Reissman et al,47 the pancreaticojejunostomy group with 18 patients underwent end-to-end anastomosis of pancreaticojejunostomy, while the ductal occlusion group with 17 patients underwent primary closure of the pancreatic duct, oversewing of the pancreatic stump, and external drainage. The ductal occlusion group had lower morbidity (56% vs 24%), decreased mortality (11% vs 0%), and a shorter hospital stay (42.2 vs 26.4 days). However, the study population was small. In the study by Tran et al,48 a total of 83 patients received pancreaticojejunostomy, while 86 patients received chemical and suture occlusion of the pancreatic duct. No significant difference was found in postoperative complications, mortality, or exocrine insufficiency. The pancreatic fistula rate was significantly higher in the ductal occlusion group (5% vs 17%). After 3 and 12 months, there were significantly more patients with diabetes mellitus in the ductal occlusion group. So far, insufficient evidence exists to show that pancreaticoenterostomy can be replaced by pancreatic ductal obliteration (evidence level 2).


Waugh and Clagett49 first used pancreaticogastrostomy in clinical practice in 1946. Pancreaticogastrostomy has gained favor in recent years as a possible means of reducing the incidence of pancreatic fistula.28,50-55 Proponents have noted several potential advantages. Pancreatic enzymes are inactivated by the gastric acidic environment, and the stomach does not contain enterokinase, which is required for conversion of trypsinogen to trypsin and subsequent activation of other proteolytic enzymes. A lack of enzyme activation may help prevent autodigestion of the anastomosis. Furthermore, the proximity of the pancreas to the posterior stomach wall allows for potentially less tension on the anastomosis. The excellent blood supply to the stomach wall is favorable to anastomotic healing, and the thickness of the stomach wall holds sutures well.

There have been 3 RCTs comparing pancreaticogastrostomy with pancreaticojejunostomy (Table 2).28,54,55 They failed to show any significant difference regarding pancreatic fistula rates, postoperative complications, or mortality. Two recent meta-analyses56,57 attempted to solve this controversy. The study by McKay et al56 analyzed 11 articles, including 1 RCT, 2 prospective nonrandomized trials, and 8 cohort studies. The study findings suggested that pancreaticogastrostomy was safer after pancreatoduodenectomy, but much of the evidence came from cohort studies. The study by Wente et al57 analyzed 16 articles, including 3 RCTs. The results indicated that all cohort studies reported superiority of pancreaticogastrostomy over pancreaticojejunostomy, most likely influenced by publication bias. In contrast, all RCTs failed to show an advantage of a particular technique, suggesting that both techniques were equally good. Based on the current evidence, pancreaticogastrostomy and pancreaticojejunostomy are equivalent in terms of perioperative outcome (evidence levels 1 and 2).


Pancreaticojejunostomy has been the most commonly used method of pancreaticoenteric anastomosis after pancreatoduodenectomy. Apart from the different positions of the jejunal loop (antecolic, retrocolic, or retromesenteric) and the other variations, such as the isolated Roux loop pancreaticojejunostomy, the anastomosis can be performed as an end-to-end anastomosis with invagination of the pancreatic stump in the jejunum or as an end-to-side anastomosis with or without duct-to-mucosa suturing. There are more variations of these 2 anastomotic techniques. An internal or external pancreatic duct stent may or may not be placed across these anastomoses. There is no firm evidence to show which technique is superior.

Isolated roux loop pancreaticojejunostomy

Separation of the pancreaticojejunal and hepaticojejunal anastomoses by an isolated Roux loop reconstruction was advocated to minimize the incidence and severity of anastomotic erosion by pancreatic juice activated by bile. Potential disadvantages are increased operating time and the need for an additional anastomosis. Several cohort studies58-60 have reported a low pancreatic fistula rate and its related mortality. In the only nonrandomized study, by Kaman et al,61 the data failed to show any significant difference in the pancreatic fistula rates (10% vs 12%) following the isolated Roux loop pancreaticojejunal reconstruction or conventional single-loop pancreaticojejunal reconstruction after pancreatoduodenectomy. Based on the limited evidence, the use of isolated Roux loop pancreaticojejunostomy cannot prevent pancreatic fistula formation (evidence levels 3b and 4).

Pancreatic duct stenting

One of the technical modifications in pancreaticoenteric anastomosis is the use of a transanastomotic stent for internal or external drainage of pancreatic secretion. The potential advantages of a pancreatic stent include diversion of pancreatic secretion from the anastomosis and facilitation for more precise placement of sutures during the anastomosis, thus protecting the pancreatic duct from suture injury and reducing the risk of iatrogenic pancreatic duct occlusion. However, complications such as obstruction of the stent leading to pancreatic fistula and migration of the stent are drawbacks with transanastomotic stenting. The number of studies62-67 on pancreatic stenting is limited, and the results are conflicting.

Internal transanastomotic stenting was reported to reduce the pancreatic fistula of pancreaticojejunal anastomosis in the cohort study by Yoshimi et al.62 However, in the nonrandomized study by Imaizumi et al64 with 168 patients, there was no significant difference in the pancreatic fistula rates between the end-to-side pancreaticojejunostomy of normal soft pancreas using the stented (internal or external) methods vs the nonstented methods (5.7% vs 6.7%). The internal pancreatic stent was evaluated by Winter et al65 in an RCT among 234 patients. The study showed that internal pancreatic duct stenting did not decrease the frequency or the severity of postoperative pancreatic fistulas. The pancreatic fistula rates in patients undergoing pancreatoduodenectomy with and without an internal pancreatic stent were 11.3% and 7.6%, respectively.

An external stent has the theoretical advantage of more complete diversion of pancreatic secretion away from the pancreaticojejunal anastomosis and prevents the activation of pancreatic enzymes by bile. The nonrandomized study by Ohwada et al66 showed equivalent outcomes for external and internal pancreatic stenting of duct-to-mucosa pancreaticojejunostomy after pancreatoduodenectomy. The RCT by Poon et al67 among 120 patients showed that the externally stented group had a significantly lower pancreatic fistula rate compared with the nonstented group (6.7% vs 20%).

Based on the current evidence, it is unclear whether drainage of the pancreatic duct with a stent can reduce the pancreatic fistula rate after pancreatoduodenectomy (evidence levels 2 and 3b).

Comparison of different anastomotic techniques

Few nonrandomized studies and RCTs have compared the various surgical techniques.68-72 Most are cohort studies.73-78 To date, no RCT has demonstrated that duct-to-mucosa anastomosis was superior to invagination anastomosis. The RCT by Bassi et al71 found no significant differences in abdominal complication rates (35% vs 38%) or pancreatic fistula rates (13% vs 15%) between 72 patients randomized to duct-to-mucosa anastomosis vs 72 patients randomized to end-to-side single-layer pancreaticojejunostomy technique. The invagination technique for the more difficult soft pancreatic stump and small pancreatic duct has been advocated by most authors,27,79 reserving the duct-to-mucosa anastomosis for the firm pancreatic stump and large pancreatic duct.

Peng et al72,80 proposed that pancreatic fistula starts at a point where a needle inadvertently penetrates a pancreatic ductule or a suture lacerates the fragile pancreatic parenchyma on suturing or tying a knot. The resultant minor leak in pancreatic juice gradually leads to gross leakage as a consequence of autodigestion around the anastomosis. Based on this hypothesis, Peng et al80 described a binding pancreaticojejunostomy technique with a pancreatic fistula rate of 0%. In brief, the stump of the jejunum was everted for a length of 3 cm. The jejunal mucosa was destroyed by applying 10% carbolic acid (subsequently rinsed with 75% alcohol and isotonic sodium chloride solution) or by diathermy coagulation. The remnant of the pancreas was dissected for a distance of greater than 3 cm from its cut edge. The cut edge of the pancreatic stump and the everted jejunum were anastomosed with 3-0 silk in a circular fashion, and care was taken to suture the jejunal mucosa only to avoid penetrating into the muscular and serosal layers. The posterior lip of the pancreatic duct should be involved in the posterior row of suture whenever technically possible. The everted jejunum was restored to its normal position to wrap over the pancreatic stump. An absorbable ligature was then looped around the jejunum, with the invaginated pancreas inside, 1.5 to 2 cm from the cut edge of the jejunum. The ligature was tied just tight enough to allow the tip of a hemostatic clamp to pass underneath the ligature. Care was taken to preserve a branch of the jejunal vessels to supply the edge of the jejunum beyond the binding ligature. This favorable outcome was further validated in an RCT demonstrating that the binding pancreaticojejunostomy after pancreaticoduodenectomy significantly decreased postoperative complications (24.5% vs 36.9% morbidity and 2.8% vs 6.3% mortality) and pancreatic fistula rates (0% vs 7.2%) and shortened the hospital stay (mean, 18.4 days vs 22.4 days) compared with the end-to-end pancreaticojejunostomy.72 Such a comprehensive procedure of pancreaticojejunostomy with this binding technique instead of the traditional suturing technique theoretically protects the pancreatic anastomosis from leakage. The main concern of the binding pancreaticojejunostomy is the difficulty in controlling the tightness of tying the binding ligature. Should the tie be too loose, watertight closure cannot be achieved, and leakage occurs. Should the tie be too tight, blood supply to the end of the pancreas can be compromised, and the pancreatic duct can be occluded.

The following 3 conditions are important for successful pancreaticojejunal anastomosis: (1) a tension-free and secured anastomosis, (2) adequate blood supply of the pancreatic stump and the jejunal end, and (3) an unobstructed passage of pancreatic secretions into the jejunum. In devising a reliable pancreaticoenteric anastomosis, a modified technique of pancreaticojejunal invagination that meets these requirements has been proposed.78 On completion of the pancreatoduodenectomy, the remnant of the pancreas was dissected for a distance of greater than 4 cm from its cut edge. Once the jejunum and pancreas were prepared, an end-to-end anastomosis was performed between the jejunum and the pancreatic stump. Two polypropylene 4-0 sutures with double needles were placed approximately 2 cm from the superior and inferior cut ends of the pancreatic stump, respectively. The long strands of the sutures of the pancreas were then passed into the open end of the jejunum and out through the wall superiorly and inferiorly, about 2 cm from the cut end of the jejunum. In this way, the pancreas was ready to be invaginated into the intestine by gentle traction of these sutures and by pulling the jejunum in the opposite direction over the pancreas. After traction, a 4-cm pancreatic stump was invaginated into the jejunum. The traction sutures were tied to avoid slippage of the pancreatic stump out of the jejunum. Healing of the pancreaticojejunal invagination was further facilitated by jejunal mucosa–controlled cauterization, which prevented the secretions from the jejunal mucosa from becoming trapped in the jejunum-pancreas interface and which induced inflammation to promote healing after invagination. Finally, the capsular edge of the transected pancreas and the free end of the jejunum were sewn circumferentially by continuous sutures (polypropylene 4-0). With our technique, the pancreaticojejunal anastomosis was invaginated to a depth of 4 cm into the jejunum, and the invagination was secured by 3 layers of anastomosis. We also adopted the technique of pancreatic neck transection based on the concept of enhancing the blood supply to the pancreatic stump as described by Strasberg and McNevin21 and by Strasberg et al.22 The pancreatic neck contains a vascular watershed. The head is supplied by the gastroduodenal and superior mesenteric arteries, which form the pancreaticoduodenal arcades. The body is supplied by branches of the splenic artery. Branches from these 2 zones of arterial supply meet in the neck of the pancreas. There are common anomalies in which the arterial supply to the neck and the adjacent part of the body comes from vessels other than the splenic artery. These vessels may be interrupted during pancreatoduodenectomy, and consequent relative ischemia may develop at the pancreaticojejunal anastomosis if the anastomosis is made at the end of the pancreatic stump. Because there is a vascular watershed at the neck of the pancreas, moving the plane of transection toward the left improves the blood supply to the pancreatic cut end. Our anastomosis between the jejunal cut end and the pancreatic stump is made 4 cm from the cut end of the pancreatic stump, ensuring an adequate blood supply to the anastomosis. We have been using this technique, which can be performed reliably with few complications, in clinical practice. In the cohort study by Chen et al78 with 52 consecutive patients, the pancreatic fistula rate after this modified invagination technique of anastomosis was 0%. The favorable results of this technique warrant further investigation in large prospective studies and RCTs. Based on the current evidence, it is unclear which pancreaticojejunostomy technique is superior (evidence levels 2, 3b, 3c, and 4).


Various pharmacologic and technical interventions have been suggested to decrease the frequency of pancreatic fistula after pancreatic resections, but the results have been controversial. More large-scale comparative studies and RCTs are required to determine the optimum pharmacologic interventions and technique of pancreaticoenteric anastomosis after pancreatoduodenectomy.

Correspondence: Wan Yee Lau, MD, FRCS, FRACS(Hon), Faculty of Medicine, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, China (

Accepted for Publication: November 26, 2008.

Author Contributions:Study concept and design: Lai. Acquisition of data: Lai, S. H. Y. Lau, and W. Y. Lau. Analysis and interpretation of data: Lai and W. Y. Lau. Drafting of the manuscript: Lai, S. H. Y. Lau, and W. Y. Lau. Critical revision of the manuscript for important intellectual content: W. Y. Lau. Statistical analysis: Lai. Administrative, technical, and material support: S. H. Y. Lau. Study supervision: W. Y. Lau.

Financial Disclosure: None reported.

Halsted  WS Contributions to the surgery of the bile passages, especially of the common bile duct.  Boston Med Surg J 1899;141645- 654Google ScholarCrossref
Kaush  W Das carcinom der papilla duodeni und seine radikale entfernung.  Beitrage zur Klinishen Chirurgie 1912;7829- 33Google Scholar
Whipple  AOParsons  WBMullins  CR Treatment of carcinoma of the ampulla of Vater.  Ann Surg 1935;102 (4) 763- 779PubMedGoogle ScholarCrossref
Whipple  AO The rationale of radical surgery for cancer of the pancreas and ampullary region.  Ann Surg 1941;114 (4) 612- 615PubMedGoogle ScholarCrossref
Whipple  AO A reminiscence: pancreaticoduodenectomy.  Rev Surg 1963;20221- 225PubMedGoogle Scholar
Matsumoto  YFujii  HMiura  K  et al.  Successful pancreatojejunal anastomosis for pancreatoduodenectomy.  Surg Gynecol Obstet 1992;175 (6) 555- 562PubMedGoogle Scholar
Cullen  JJSarr  MGIlstrup  DM Pancreatic anastomotic leak after pancreaticoduodenectomy: incidence, significance, and management.  Am J Surg 1994;168 (4) 295- 298PubMedGoogle ScholarCrossref
Yeo  CJCameron  JLSohn  TA  et al.  Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes.  Ann Surg 1997;226 (3) 248- 260PubMedGoogle ScholarCrossref
Büchler  MWFriess  HWagner  MKulli  CWagener  VZ'Graggen  K Pancreatic fistula after pancreatic head resection.  Br J Surg 2000;87 (7) 883- 889PubMedGoogle ScholarCrossref
Schmidt  CMPowell  ESYiannoutsos  CT  et al.  Pancreaticoduodenectomy: a 20-year experience in 516 patients.  Arch Surg 2004;139 (7) 718- 727PubMedGoogle ScholarCrossref
Cameron  JLRiall  TSColeman  JBelcher  KA One thousand consecutive pancreaticoduodenectomies.  Ann Surg 2006;244 (1) 10- 15PubMedGoogle ScholarCrossref
Bassi  CDervenis  CButturini  G  et al. International Study Group on Pancreatic Fistula Definition, Postoperative pancreatic fistula: an international study group (ISGPF) definition.  Surgery 2005;138 (1) 8- 13PubMedGoogle ScholarCrossref
Neoptolemos  JPRussell  RCBramhall  STheis  BUK Pancreatic Cancer Group, Low mortality following resection for pancreatic and periampullary tumours in 1026 patients: UK survey of specialist pancreatic units.  Br J Surg 1997;84 (10) 1370- 1376PubMedGoogle ScholarCrossref
Gordon  TABowman  HMBass  EB  et al.  Complex gastrointestinal surgery: impact of provider experience on clinical and economic outcomes.  J Am Coll Surg 1999;189 (1) 46- 56PubMedGoogle ScholarCrossref
Gouma  DJvan Geenen  RCvan Gulik  TM  et al.  Rates of complications and death after pancreaticoduodenectomy: risk factors and the impact of hospital volume.  Ann Surg 2000;232 (6) 786- 795PubMedGoogle ScholarCrossref
McPhee  JTHill  JSWhalen  GF  et al.  Perioperative mortality for pancreatectomy: a national perspective.  Ann Surg 2007;246 (2) 246- 253PubMedGoogle ScholarCrossref
Topal  BVan de Sande  SFieuws  SPenninckx  F Effect of centralization of pancreaticoduodenectomy on nationwide hospital mortality and length of stay.  Br J Surg 2007;94 (11) 1377- 1381PubMedGoogle ScholarCrossref
Balzano  GZerbi  ACapretti  GRocchetti  SCapitanio  VDi Carlo  V Effect of hospital volume on outcome of pancreaticoduodenectomy in Italy.  Br J Surg 2008;95 (3) 357- 362PubMedGoogle ScholarCrossref
National Health and Medical Research Council, How to review the evidence: systematic identification and review of the scientific literature.  Canberra, Australia National Health and Medical Research Council2000;
Bartoli  FGArnone  GBRavera  GBachi  V Pancreatic fistula and relative mortality in malignant disease after pancreaticoduodenectomy: review and statistical meta-analysis regarding 15 years of literature.  Anticancer Res 1991;11 (5) 1831- 1848PubMedGoogle Scholar
Strasberg  SM McNevin  MS Results of a technique of pancreaticojejunostomy that optimizes blood supply to the pancreas.  J Am Coll Surg 1998;187 (6) 591- 596PubMedGoogle ScholarCrossref
Strasberg  SMDrebin  JAMokadam  NA  et al.  Prospective trial of a blood supply–based technique of pancreaticojejunostomy: effect on anastomotic failure in the Whipple procedure.  J Am Coll Surg 2002;194 (6) 746- 760PubMedGoogle ScholarCrossref
Hashimoto  NOhyanagi  H Pancreatic juice output and amylase level in the drainage fluid after pancreatoduodenectomy in relation to leakage.  Hepatogastroenterology 2002;49 (44) 553- 555PubMedGoogle Scholar
Shyr  YMSu  CHWu  CWLui  WY Does drainage fluid amylase reflect pancreatic leakage after pancreaticoduodenectomy?  World J Surg 2003;27 (5) 606- 610PubMedGoogle ScholarCrossref
Shrikhande  SVQureshi  SSRajneesh  NShukla  PJ Pancreatic anastomoses after pancreaticoduodenectomy: do we need further studies?  World J Surg 2005;29 (12) 1642- 1649PubMedGoogle ScholarCrossref
Molinari  EBassi  CSalvia  R  et al.  Amylase value in drains after pancreatic resection as predictive factor of postoperative pancreatic fistula: results of a prospective study in 137 patients.  Ann Surg 2007;246 (2) 281- 287PubMedGoogle ScholarCrossref
Marcus  SGCohen  HRanson  JH Optimal management of the pancreatic remnant after pancreaticoduodenectomy.  Ann Surg 1995;221 (6) 635- 648PubMedGoogle ScholarCrossref
Yeo  CJCameron  JLMaher  MM  et al.  A prospective randomized trial of pancreaticogastrostomy versus pancreaticojejunostomy after pancreaticoduodenectomy.  Ann Surg 1995;222 (4) 580- 592PubMedGoogle ScholarCrossref
van Berge Henegouwen  MIDe Wit  LTVan Gulik  TMObertop  HGouma  DJ Incidence, risk factors, and treatment of pancreatic leakage after pancreaticoduodenectomy: drainage versus resection of the pancreatic remnant.  J Am Coll Surg 1997;185 (1) 18- 24PubMedGoogle ScholarCrossref
Sato  NYamaguchi  KChijiiwa  KTanaka  M Risk analysis of pancreatic fistula after pancreatic head resection.  Arch Surg 1998;133 (10) 1094- 1098PubMedGoogle ScholarCrossref
Popiela  TKedra  BSierzega  MGurda  A Risk factors of pancreatic fistula following pancreaticoduodenectomy for periampullary cancer.  Hepatogastroenterology 2004;51 (59) 1484- 1488PubMedGoogle Scholar
Yang  YMTian  XDZhuang  YWang  WMWan  YLHuang  YT Risk factors of pancreatic leakage after pancreaticoduodenectomy.  World J Gastroenterol 2005;11 (16) 2456- 2461PubMedGoogle Scholar
Murakami  YUemura  KHayasidani  Y  et al.  A soft pancreatic remnant is associated with increased drain fluid pancreatic amylase and serum CRP levels following pancreatoduodenectomy.  J Gastrointest Surg 2008;12 (1) 51- 56PubMedGoogle ScholarCrossref
Büchler  MFriess  HKlempa  I  et al.  Role of octreotide in the prevention of postoperative complications following pancreatic resection.  Am J Surg 1992;163 (1) 125- 131PubMedGoogle ScholarCrossref
Pederzoli  PBassi  CFalconi  MCamboni  MGItalian Study Group, Efficacy of octreotide in the prevention of complications of elective pancreatic surgery.  Br J Surg 1994;81 (2) 265- 269PubMedGoogle ScholarCrossref
Montorsi  MZago  MMosca  F  et al.  Efficacy of octreotide in the prevention of pancreatic fistula after elective pancreatic resections: a prospective, controlled, randomized clinical trial.  Surgery 1995;117 (1) 26- 31PubMedGoogle ScholarCrossref
Friess  HBeger  HGSulkowski  U  et al.  Randomized controlled multicentre study of the prevention of complications by octreotide in patients undergoing surgery for chronic pancreatitis [published correction appears in Br J Surg. 1996;83(1):126].  Br J Surg 1995;82 (9) 1270- 1273PubMedGoogle ScholarCrossref
Lowy  AMLee  JEPisters  PW  et al.  Prospective, randomized trial of octreotide to prevent pancreatic fistula after pancreaticoduodenectomy for malignant disease.  Ann Surg 1997;226 (5) 632- 641PubMedGoogle ScholarCrossref
Yeo  CJCameron  JLLillemoe  KD  et al.  Does prophylactic octreotide decrease the rates of pancreatic fistula and other complications after pancreaticoduodenectomy? results of a prospective randomized placebo-controlled trial.  Ann Surg 2000;232 (3) 419- 429PubMedGoogle ScholarCrossref
Gouillat  CChipponi  JBaulieux  JPartensky  CSaric  JGayet  B Randomized controlled multicentre trial of somatostatin infusion after pancreaticoduodenectomy.  Br J Surg 2001;88 (11) 1456- 1462PubMedGoogle ScholarCrossref
Sarr  MGPancreatic Surgery Group, The potent somatostatin analogue vapreotide does not decrease pancreas-specific complications after elective pancreatectomy: a prospective, multicenter, double-blinded, randomized, placebo-controlled trial.  J Am Coll Surg 2003;196 (4) 556- 565PubMedGoogle ScholarCrossref
Shan  YSSy  EDLin  PW Role of somatostatin in the prevention of pancreatic stump–related morbidity following elective pancreaticoduodenectomy in high-risk patients and elimination of surgeon-related factors: prospective, randomized, controlled trial.  World J Surg 2003;27 (6) 709- 714PubMedGoogle ScholarCrossref
Suc  BMsika  SPiccinini  M  et al. French Associations for Surgical Research, Octreotide in the prevention of intra-abdominal complications following elective pancreatic resection: a prospective, multicenter randomized controlled trial.  Arch Surg 2004;139 (3) 288- 295PubMedGoogle ScholarCrossref
Hesse  UJDeDecker  CHoutmeyers  P  et al.  Prospectively randomized trial using perioperative low-dose octreotide to prevent organ-related and general complications after pancreatic surgery and pancreatico-jejunostomy [published correction appears in World J Surg. 2006;30(4):641].  World J Surg 2005;29 (10) 1325- 1328PubMedGoogle ScholarCrossref
Connor  SAlexakis  NGarden  OJLeandros  EBramis  JWigmore  SJ Meta-analysis of the value of somatostatin and its analogues in reducing complications associated with pancreatic surgery.  Br J Surg 2005;92 (9) 1059- 1067PubMedGoogle ScholarCrossref
Alghamdi  AAJawas  AMHart  RS Use of octreotide for the prevention of pancreatic fistula after elective pancreatic surgery: a systematic review and meta-analysis.  Can J Surg 2007;50 (6) 459- 466PubMedGoogle Scholar
Reissman  PPerry  YCuenca  A  et al.  Pancreaticojejunostomy versus controlled pancreaticocutaneous fistula in pancreaticoduodenectomy for periampullary carcinoma.  Am J Surg 1995;169 (6) 585- 588PubMedGoogle ScholarCrossref
Tran  KVan Eijck  CDi Carlo  V  et al.  Occlusion of the pancreatic duct versus pancreaticojejunostomy: a prospective randomized trial.  Ann Surg 2002;236 (4) 422- 428PubMedGoogle ScholarCrossref
Waugh  JMClagett  OT Resection of the duodenum and head of pancreas for carcinoma: an analysis of thirty cases.  Surgery 1946;20224- 232Google Scholar
Sauvanet  ABelghiti  JPanis  Y  et al.  Pancreaticogastrostomy after pancreatoduodenectomy.  HPB Surg 1992;6 (2) 91- 98PubMedGoogle ScholarCrossref
Takano  SIto  YWatanabe  YYokoyama  TKubota  NIwai  S Pancreaticojejunostomy versus pancreaticogastrostomy in reconstruction following pancreaticoduodenectomy.  Br J Surg 2000;87 (4) 423- 427PubMedGoogle ScholarCrossref
Aranha  GVHodul  PGolts  EOh  DPickleman  JCreech  S A comparison of pancreaticogastrostomy and pancreaticojejunostomy following pancreaticoduodenectomy.  J Gastrointest Surg 2003;7 (5) 672- 682PubMedGoogle ScholarCrossref
Arnaud  JPTuech  JJCervi  CBergamaschi  R Pancreaticogastrostomy compared with pancreaticojejunostomy after pancreaticoduodenectomy.  Eur J Surg 1999;165 (4) 357- 362PubMedGoogle ScholarCrossref
Duffas  JPSuc  BMsika  S  et al. French Associations for Research in Surgery, A controlled randomized multicenter trial of pancreatogastrostomy or pancreatojejunostomy after pancreatoduodenectomy.  Am J Surg 2005;189 (6) 720- 729PubMedGoogle ScholarCrossref
Bassi  CFalconi  MMolinari  E  et al.  Reconstruction by pancreaticojejunostomy versus pancreaticogastrostomy following pancreatectomy: results of a comparative study.  Ann Surg 2005;242 (6) 767- 773PubMedGoogle ScholarCrossref
McKay  AMackenzie  SSutherland  FR  et al.  Meta-analysis of pancreaticojejunostomy versus pancreaticogastrostomy reconstruction after pancreaticoduodenectomy.  Br J Surg 2006;93 (8) 929- 936PubMedGoogle ScholarCrossref
Wente  MNShrikhande  SVMüller  MW  et al.  Pancreaticojejunostomy versus pancreaticogastrostomy: systematic review and meta-analysis.  Am J Surg 2007;193 (2) 171- 183PubMedGoogle ScholarCrossref
Kingsnorth  AN Safety and function of isolated Roux loop pancreaticojejunostomy after Whipple's pancreaticoduodenectomy.  Ann R Coll Surg Engl 1994;76 (3) 175- 179PubMedGoogle Scholar
Khan  AWAgarwal  AKDavidson  BR Isolated Roux loop duct-to-mucosa pancreaticojejunostomy avoids pancreatic leaks in pancreaticoduodenectomy.  Dig Surg 2002;19 (3) 199- 204PubMedGoogle ScholarCrossref
Sutton  CDGarcea  GWhite  SA  et al.  Isolated Roux-loop pancreaticojejunostomy: a series of 61 patients with zero postoperative pancreaticoenteric leaks.  J Gastrointest Surg 2004;8 (6) 701- 705PubMedGoogle ScholarCrossref
Kaman  LSanyal  SBehera  ASingh  RKatariya  RN Isolated Roux loop pancreaticojejunostomy vs single loop pancreaticojejunostomy after pancreaticoduodenectomy.  Int J Surg 2008;6 (4) 306- 310PubMedGoogle ScholarCrossref
Yoshimi  FOno  HAsato  Y  et al.  Internal stenting of the hepaticojejunostomy and pancreaticojejunostomy in patients undergoing pancreatoduodenectomy to promote earlier discharge from hospital.  Surg Today 1996;26 (8) 665- 667PubMedGoogle ScholarCrossref
Roder  JDStein  HJBöttcher  KABusch  RHeidecke  CDSiewert  JR Stented versus nonstented pancreaticojejunostomy after pancreatoduodenectomy: a prospective study.  Ann Surg 1999;229 (1) 41- 48PubMedGoogle ScholarCrossref
Imaizumi  THatori  TTobita  KFukuda  ATakasaki  KMakuuchi  H Pancreaticojejunostomy using duct-to-mucosa anastomosis without a stenting tube.  J Hepatobiliary Pancreat Surg 2006;13 (3) 194- 201PubMedGoogle ScholarCrossref
Winter  JMCameron  JLCampbell  KA  et al.  Does pancreatic duct stenting decrease the rate of pancreatic fistula following pancreaticoduodenectomy? results of a prospective randomized trial.  J Gastrointest Surg 2006;10 (9) 1280- 1290PubMedGoogle ScholarCrossref
Ohwada  STanahashi  YOgawa  T  et al.  In situ vs ex situ pancreatic duct stents of duct-to-mucosa pancreaticojejunostomy after pancreaticoduodenectomy with Billroth I–type reconstruction.  Arch Surg 2002;137 (11) 1289- 1293PubMedGoogle ScholarCrossref
Poon  RTFan  STLo  CM  et al.  External drainage of pancreatic duct with a stent to reduce leakage rate of pancreaticojejunostomy after pancreaticoduodenectomy: a prospective randomized trial.  Ann Surg 2007;246 (3) 425- 435PubMedGoogle ScholarCrossref
Hosotani  RDoi  RImamura  M Duct-to-mucosa pancreaticojejunostomy reduces the risk of pancreatic leakage after pancreatoduodenectomy.  World J Surg 2002;26 (1) 99- 104PubMedGoogle ScholarCrossref
Langrehr  JMBahra  MJacob  DGlanemann  MNeuhaus  P Prospective randomized comparison between a new mattress technique and Cattell (duct-to-mucosa) pancreaticojejunostomy for pancreatic resection.  World J Surg 2005;29 (9) 1111- 1121PubMedGoogle ScholarCrossref
Lee  SEYang  SHJang  JYKim  SW Pancreatic fistula after pancreaticoduodenectomy: a comparison between the two pancreaticojejunostomy methods for approximating the pancreatic parenchyma to the jejunal seromuscular layer: interrupted vs continuous stitches.  World J Gastroenterol 2007;13 (40) 5351- 5356PubMedGoogle Scholar
Bassi  CFalconi  MMolinari  E  et al.  Duct-to-mucosa versus end-to-side pancreaticojejunostomy reconstruction after pancreaticoduodenectomy: results of a prospective randomized trial.  Surgery 2003;134 (5) 766- 771PubMedGoogle ScholarCrossref
Peng  SYWang  JWLau  WY  et al.  Conventional versus binding pancreaticojejunostomy after pancreaticoduodenectomy: a prospective randomized trial.  Ann Surg 2007;245 (5) 692- 698PubMedGoogle ScholarCrossref
Z’graggen  KUhl  WFriess  H  et al.  How to do a safe pancreatic anastomosis.  J Hepatobiliary Pancreat Surg 2002;9 (6) 733- 737PubMedGoogle ScholarCrossref
Ibrahim  STay  KHLaunois  BTa  NC Triple-layer duct-to-mucosa pancreaticojejunostomy after pancreaticoduodenectomy.  Dig Surg 2006;23 (5-6) 296- 302PubMedGoogle ScholarCrossref
Hayashibe  AKameyama  M The clinical results of duct-to-mucosa pancreaticojejunostomy after pancreaticoduodenectomy in consecutive 55 cases.  Pancreas 2007;35 (3) 273- 275PubMedGoogle ScholarCrossref
Murakami  YUemura  KHayashidani  Y  et al.  No mortality after 150 consecutive pancreatoduodenectomies with duct-to-mucosa pancreaticogastrostomy.  J Surg Oncol 2008;97 (3) 205- 209PubMedGoogle ScholarCrossref
Peng  SMou  YCai  XPeng  C Binding pancreaticojejunostomy is a new technique to minimize leakage.  Am J Surg 2002;183 (3) 283- 285PubMedGoogle ScholarCrossref
Chen  HWLai  ECSu  SYCai  YFZhen  ZJLau  WY Modified technique of pancreatico-jejunal anastomosis with invagination following pancreaticoduodenectomy: a cohort study.  World J Surg 2008;32 (12) 2695- 2700PubMedGoogle ScholarCrossref
Suzuki  YFujino  YTanioka  Y  et al.  Selection of pancreaticojejunostomy techniques according to pancreatic texture and duct size.  Arch Surg 2002;137 (9) 1044- 1048PubMedGoogle ScholarCrossref
Peng  SYMou  YPLiu  YB  et al.  Binding pancreaticojejunostomy: 150 consecutive cases without leakage.  J Gastrointest Surg 2003;7 (7) 898- 900PubMedGoogle ScholarCrossref