Department of Hepatobiliary & Pancreatic Surgery and Liver Transplantation, Anhui Medical University, Anhui, China
Received date: February 15, 2017; Accepted date: February 20, 2017; Published date: February 27, 2017
Citation: Mazid MA, Akter GS, Ye ZH, Geng X, Liu F, et al. Minimizing Occurrence of Pancreatic Fistula during Pancreatoduodenectomy (Pd) Procedure: An Update. Journal of Surgery [Jurnalul de chirurgie]. 2017; 13(1): 11-16 DOI: 10.7438/1584-9341-13-1-3
Copyright: © 2017 Mazid MA, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Surgery
Background: Pancreatic fistula (PF) is the most widely recognized complication of pancreaticoduodenectomy (PD) with diversely reported occurrence rates. Present review focusses on dissecting the surgical treatment modalities that leads to development of PF. Methods: A retrospective study with the use of hospital database as cases and controls was carried out. Data were tabulated and subjected to strong statistical analysis and inferences were drawn. Results: As observed the occurrence of PF did not differ in terms of mean age, sex, surgical timings to the procedure, anastomosis techniques or vascular resection. Conclusion: The surgical approach for PF is related with a higher mortality and morbidity. There is no preferred method of performing pancreatectomy as any procedure can give rise to same mortality rates and risk of endocrine deficiency. In instances of muddled PF, radiological or surgical conservative treatment is needed and surgically duct to mucosal double layered anastomosis have been successful in reducing the PF rates and its validation is still awaited from the trials.
Pancreatoduodenectomy; Duct-to-duct (DTD) Pancreatic fistula; Pancreaticojejunostomy
What is PD? Pancreatico-duodenectomy is a surgical term applied to the procedure of removal of various extents of pancreas and duodenum. Although the term itself is incomplete as to the extent of removal of organs as it also encompasses removal of common hepatic, common bile duct and gall bladder as well as shown in (Figure 1). Chiefly this procedure is performed in cases of pancreatic head cancer both as a therapeutic and palliative treatment. It’s one of the surgical procedures that demands optimal surgical skills and vigilance as with the success also, the patient can succumb into great deal of morbidity, chiefly pertaining to the digestive system.
Recent literature recommends that many elements impact pancreatic leakage after PD, including sex, age, jaundice, operative time, intraoperative blood loss, pancreaticojejunal anastomotic strategy, surface of the remnant pancreas, pancreatic duct size, utilization of somatostatin, and specialist's experience. Different methodologies have been utilized to diminish the rate of PF including pharmacological controls and refinements and alterations in surgical systems, which are evaluated here. We hope that the findings presented in this article will be informative and help the scholars to be updated with the recent clinical studies conducted as well as what to expect in future findings.
For this review, the PubMed database was searched for articles concerning pancreatic fistula and the surgical procedures employed during PD, published in English before July 2016. We used the search terms “Pancreatic fistula” and “Pancreatico-duodenectomy”. Clinical study was considered if they evaluated the association of pancreatic fistula rates the pathogenesis, pathological features, and surgical methods in accordance to author’s judgment.
Common methods employed in PD: As depicted in Figure 1, there are chiefly 3 types of surgeries performed in PD. Due to the demand of sophisticated surgical facility and technical skills required, the classical procedure of PD is commonly deployed over the world and in low volume centers.
Complications with PD: From the very start of practice with this procedure, besides non surgery associated complications, various complications have been observed as depicted in (Table I).
Table I: Overall complications associated with PD.
|Local Complications||Systemic Complications|
|Wound infection||Myocardial infarction|
|Delayed gastric emptying||Stroke|
|Acute necrotizing pancreatitis|
Recently the operative mortality after PD has essentially declined to 3 to 5%, while the rate of postoperative morbidity stays high running from 30% to 65%. The absolute most noteworthy reason for morbidity and mortality after PD is the improvement of pancreatic hole and fistula. The leakage rate as per late reports shifts from 0% to 25% contingent upon the definition used [1-3]. Stomach sore and discharge are regular sequelae of pancreatic anastomotic leakage which have frequently been connected with a death rate of at least 40%. Pancreatic fistula (PF) consequently has been one of the real entanglements demoralizing specialists from performing PD.
There has been varying opinions as to the chief and major complications associated with PD. However if we consider the surgical procedure strictly and focus on the organs involved in the procedure, the occurrence of pancreatic fistula stands tall as compared to the occurrence of wound infection. The latter can be associated with any surgical and most medical conditions while pancreatic fistula occurs only with the procedure of PD.
PF Is defined as an abnormal connection between the pancreas and adjacent or distant organs, structures, or spaces. Historically we can see that in order to define and categorize PF, difference in parameters exists as to the amount of amylase rich pancreatic rich fluid (>10 ml/ day, >50 ml/day) as well as in the index time of occurrence (at day 5 or >5 days, at day 11 or for > 11 days) [1-4]. However there has been no consensus in regards to the threshold amount of amylase level as a definition of PF. Similarly the available definitions have been largely based in the daily volume of effluent criteria and the occurrence of fistula development. Adhering to the definitions provided in these studies, the incidence range of PF ranges from 10-29% . To address this, the International Study Group for Pancreatic Fistulas (ISGPF) that comprised of 37 surgeons from various 15 countries compiled, extended and standardized the postoperative PF definition as an external fistula with a drain output of any measurable volume after postoperative day 3 with an amylase level greater than three times the upper limit of the normal serum value. Furthermore, considering the significance of the clinical in hospital course as well as the final outcome, ISGPF also graded the PF into various grades as depicted in (Table II).
Table II: Pancreatic fistula grading system based on clinical events as well as the final outcome.
|No Fistula||Grad A||Grad B||Grad C|
|Drain Amylase level||<3 times normal serum amylase||>3 times normal serum amylase||>3 times normal serum amylase||>3 times normal serum amylase|
|Signs of infection||No||No||Yes||Yes|
|Death related to fistula||No||No||No||Yes|
|Persistent drainage (>3 weeks)||No||No||Usually Yes||Yes|
|Clinical conditions||Well||Well||Often Well||Ill appearing|
|US/CT if obtained||Negative||Negative||Negative/positive||Positive|
Source: C. Bassi, C. Dervenis, G. Butturini et al., “International study group on pancreatic fistula definition. Post-operative pancreatic fistula: an international study group (ISGPF) definition,” Surgery, vol. 138, no. 1, pp. 8–13, 2005
We can see that the above criteria consider 9 clinical events. There have been disagreements to theses set 9 criteria by some scholars and some even have attempted to include other intra-abdominal events like peritonitis and hemorrhage that can result due to PF while few others have tried to include consequences of problems that set in with newer anastomotic pathway .
The outcome of PF is expanded danger of morbidity, mortality, and longer clinic stay and cost. Among the quantity of arrangement as of late distributed, the revealed occurrence of PF taking after pancreaticoduodenectomy extended from 6% to 14% and the announced mortality from 1.4% [1-11]. In addition, likewise, PF is connected with different non fistulous complications, especially delayed gastric purging, ileus, wound contamination, intra-stomach abscess, pancreatitis, discharge, and sepsis. The doctor's facility expenses and rate of reoperation and clinic readmission are essentially increased [3-8]. Table III outlines the risk factors that are associated with PF occurrence. Some of the validations received are also depicted.
Table III: Risk factors for PF occurrence.
|Parameters||Predisposition to PF (studies)|
|Patient related||Age||>70 years old|
|Jaundice and Creatinine clearance||Impairment of anastomosis healing|
|Coronary artery disease||4 fold increased likelihood|
|Disease Related||Pancreatic pathology|
|Pancreatic texture||Soft texture has 22.6% more chances of PF and has 10% more chances of having PF|
|Pancreatic duct size||Dilated ducts or ducts sixe of >3 mm|
|Pancreatic juice output||Increased pancreatic juice|
|Procedure related||Operative time|
|Imperative blood loss||Blood loss of >1.5 litres|
|Resection type||Central vs. Distal pancreatectomy|
Some of unfavorable circumstances for pancreatic fistula development are:
• Intermediate, hard or fibrotic pancreatic tissue
• Pancreatic duct size<3 mm
Since our focus of study was to discuss on the procedures and studies associated with reducing rates of PF, these are discussed in detail.
Central pancreatectomy bears the highest rate of PF formation (20% to 63%)  as compared to distal pancreatectomy 5% . One of the studies has also revealed this fact as true where the incidence of PF was 83% for the central and 13% for the distal pancreatectomy while the overall PF occurrence was 16% (Grades A, B, C of ISGPE) in patients undergoing pancreaticoduodenectomy . Moreover, the patients undergoing former procedure also required reoperations and aggressive management in the ICU setting [1,4-7] and had longer hospital stay and required discharge to rehabilitation facilities  as compared to distal pancreactomy patients.
Delayed waste of intra-stomach accumulations of over 3 weeks and numerous doctors’ facility readmissions (for the most part for image guided percutaneous) are more probable following leaks related after PD [1-8]. The dominating elements connected with the break were expanded weight, higher American Culture of Anesthesiologists score, and blood loss more prominent than 1 L, expanded operation time, diminished albumin level, and sutured conclusion of the stump without the main duct ligation.
Pancreatic duct occlusion
Various techniques either by suture ligation or by application of non-absorbable or re absorbable glue have been applied to occlude the pancreatic duct that is responsible for leakage of pancreatic duct. Some studies have found total occlusion to be effective in terms of lower morbidity , decreased mortality and shorter hospital however other studies have demonstrated indifferent results  between PJ and ductal occlusion. Moreover the latter studies also revealed a higher rate of PF occurrence.
Diverting the pancreatic remnant into the stomach had been tried by Waugh and Clagett in 1946. This procedure had advantages in terms of elimination of digestive properties of pancreatic juice by the gastric acid and reducing the occurrence of PF. Numerous studies have had supported this fact and as a result this procedure received 3 RCT’s, which however failed to demonstrated significant outcome in terms of PF rates, Post-operative complications and mortality [4,13]. Although besides this all the meta analysis done, series published, meta-analysis done have advocated superiority of PG to PD, all of them have suffered publication bias and lack reproducibility in prospective studies [2,3]. Till today both the procedures stand equally efficacious in terms of post-operative outcomes (evidence level 1 and 2).
This procedure has been practiced for long has been remarkable in reducing the rate of PF formation to 10% (range 2- 19%)  during the last 3 decades. PJ is a procedure where the pancreatic stump is connected to the jejunum (that has rich blood supply for healing of the anastomosis and has mobile mesentry). The anastomosis can be performed by direct end to end invagination of the pancreatic remnant with the pancreatic duct into the jejunum or by performing duct- mucosa anastomosis where the pancreatic duct opens into the mucosal surface of the jejunum. The later procedure limits pouring pancreatic juice into the jejunum and theoretically curtails pancreatic juice associated complications. This duct to mucosa procedure is difficult, surgically demanding yet some scholars have opined it to be safer than invagination anastomosis [2,4]. However, some studies have indeed outlined the role of risk factors such as pancreatic texture, size of the pancreatic duct as important in determining the success of ductto –mucosa technique measured by reduction of PF, RCT performed consecutively revealed indifferent results when duct-to-mucosa was compared with invagination technique.
Isolated roux loop pancreaticojejunostomy
It was also found that the combined anastomosis of PJ and HJ performed in the classical procedure lead to release of bile that helped activate the pancreatic juice and propensity for added digestive damage to anastomosis. With the procedure of isolated Roux Loop, separation of these two anastomosis points was done. Some of the cohort studies demonstrated exciting results of reduction in PF occurrence and mortality, which however received a setback when a following nonrandomized study revealed indifferent results . Thus for now, we can say that isolated Roux loop PG is not associated with lower PF formation (evidence level 3b and 4).
a. Utilization of working magnifying instrument for reproduction: Some have detailed notably decreased rate of PF with the working magnifying lens contrasted with operating loupes . Operating loupes have been utilized by numerous specialists to permit exact recreation of pancreatic anastomosis.
b. Anastomotic site: An idea of vascular watershed in the pancreatic neck and its part in ischemia of the cut surface of pancreatic leftover has been proposed by Strasberg et al.  them. In view of this idea, the blood supply at the cut surface of the pancreas is assessed in the methods and if important the pancreas is decreased 1.5 cm to 2 cm to enhance the blood supply (38%).
d. Stenting of the pancreatic duct: However complications, for example, block of the stent prompting to pancreatic fistula and relocation of the stent are disadvantages with transanastomotic stenting. The quantity of studies on pancreatic stenting is restricted and the outcomes are clashing. However in the nonrandomized think about by Imaizumi et al. with 168 patients, there was no huge contrast in the pancreatic fistula rates between end-to-side pancreaticojejunostomy of ordinary delicate pancreas utilizing stented (inward or outside) technique versus non stented strategies (5.7% vs. 6.7%) . However a few reviews have demonstrated that the inside pancreatic duct stenting did not diminish the recurrence or the seriousness of the postoperative fistulas, between PD with or without an inner pancreatic stent (11.3 vs. 7.6%, respectively) .
e. Perioperative somatostatin: The method of reasoning of its utilization taking after PD is that by diminishing the volume of pancreatic emission, the pancreatic fistula rate would be diminished due to which the pancreaticoenteric anastomosis would heal better. Perioperative utilization of somatostatin simple and its useful impacts has been seen in a portion of the reviews directed in Europe and Asia. Notwithstanding, contentions have existed with a portion of the reviews [2-4].
With these mixed and inconsistent results, numerous preoperative, per-operative and post procedural strategies have gained into clinical trials as depicted in (Table IV). Table IV shows a list of trials that are aimed at reducing PF and depicts the mechanisms manipulated. Among the trials that have been completed and results available, preoperative interventions as well as post-operative strategies have been significant in reducing the occurrence of PF. Manipulation of surgical methods have not been impressive. Yet newer strategies that focus in the surgical technique, use of modified staples for anastomotic closure and manipulating the drainage system are the studies we hope to yield some critical information in the future.
Table IV: Studies pertaining to PF within the last 5 years. Note: the trials whose chief objective was other than reducing PF occurrence are excluded.
|Ultrasound Elastography for Prediction of Postoperative Pancreatic FistulaNCT02589379||University of Zurich||Diagnostic Recruiting|
|Does Reinforcement of the Staple Line in Left Pancreatectomy Reduce the Rate of Pancreatic Fistula? NCT02149446||Karolinska University Hospital, Sweden||Suture Recruiting|
|Does Post Operative Pancreatic Fistula, After Left Sided Resections, Heal Faster After the Introduction of a Pancreatic Stent? NCT02220010||Karolinska University Hospital, Sweden||Stent Recruiting|
|Predictive Risk Factors for Pancreatic Fistula Grade C After Pancreaticoduodenectomy NCT02322424||Wakayama Medical University, Japan||Risk Factor Recruiting|
|Different Stapler Cartridge For Pancreatic Stump Texture To Prevent Pancreatic FistulaNCT02790333||Yi-Ping Mou, Zhejiang Provincial People's Hospital, HangZhou, Zhejiang, China, 310014||Suture Recruiting|
|Route of Nutritional Support for Pancreatic FistulaNCT01755260||National Taiwan University Hospital||Nutrition Recruiting|
|Pancreaticoduodenectomy With or Without Braun Enteroenterostomy: Comparison of Postoperative Pancreatic Fistula and Delayed Gastric Emptying NCT01481753||Johns Hopkins University||Procedural Recruiting|
|Prospective Trial Evaluating the Effect of Closed Suction Drainage Versus Straight Drainage After Distal Pancreatectomy NCT02343302||Johns Hopkins University||Drainage Recruiting|
|An Evaluation of a New Technique Utilizing a Biologic Glue and Tissue Patch to Seal the Cut Edge of the Pancreas Following Removal of the Tail of the Pancreas NCT00889213||Thomas Jefferson University||Procedural Recruiting|
|One-layer Versus Two-layer Duct-to-mucosa Pancreaticojejunostomy After Pancreaticoduodenectomy NCT02511951||The Second Hospital of Anhui Medical University||Procedural Recruiting|
|Drains in Pancreatic Surgery (DRAPA) NCT01988519||University Hospital Hradec Kralove||Drainage Recruiting|
|A Prospective, Multi-center Trial for Reinforced Staple During Distal Pancreatectomy NCT02270554||Wakayama Medical University||Suture Recruiting|
|External Drainage Versus Internal Drainage of Pancreatic Duct With a Stent After Pancreaticoduodenectomy (EDIDPD) NCT01634971||Tianjin Medical University Cancer Institute and Hospital||Drainage Recruiting|
|Early Versus Late Drain Removal After Pancreatectomy: A Randomized Prospective TrialNCT02230436||Peking Union Medical College Hospital||Drainage Recruiting|
|A Randomized Trial of Two Surgical Techniques for Pancreaticojejunostomy in Patients Undergoing Pancreaticoduodenectomy NCT00359320||Thomas Jefferson University||Procedural Recruiting|
|Enteral Nutrition in the Treatment of Pancreatic Fistulas - A Prospective Study NCT01025414||Jagiellonian University, Poland||Enteral nutrition is associated with significantly higher closure rates and shorter time to closure of postoperative pancreatic fistula.|
|Use of Polyethylene Glycolic Acid or Tachocomb to Prevent Pancreatic Fistula Following Distal Pancreatectomy NCT01550406||Seoul National University Hospital||Study results awaited|
|Role of Octreotide in Preventing Pancreatic Fistula After Pancreaticoduodenectomy (PD) in Patients With Soft Pancreas, NCT01301222||PVS Memorial Hospital, Kochi, Kerala, India, 682017||Not preventive|
|Reduced Pancreatic Fistula Rate Following Pancreaticoduodenectomy: Trial on Pancreaticogastrostomy Versus Pancreaticojejunostomy NCT00830778||Baki Topal, Belgium||Results awaited|
|13 Trial Assessing Roux-en-Y Anastomosis of the Pancreatic Stump to Prevent Pancreatic Fistula Following Distal Pancreatectomy|
|Stereotactic Radiation to Decrease Pancreatic Secretions NCT01656486||Carolinas Healthcare System, USA||Results awaited|
|17 Comparison of Feasibility Between Internal and External Pancreatic Drainage in Pancreaticoduodenectomy||Active|
|Closed Suction Drainage and Natural Drainage of the Pancreatic Duct in Pancreaticojejunostomy NCT00679952||Seoul National University Hospital||Completed has results|
|External Pancreatic Duct Stent After Pancreaticoduodenectomy NCT01068886||University Hospital, Angers||Decreases PF rate|
|Isolated Roux Loop Pancreaticojejunostomy Versus Pancreaticogastrostomy After Pancreaticoduodenectomy NCT01859806||Mansoura University||Both procedures are not associated with lower incidence of post operative PF occurence|
|Effects of Intraarterial Octreotide on Pancreatic Texture NCT01400100||St. Josef Hospital Bochum||A single blous did not deliver clinically significant pancreatic hardness|
|Randomized Trial of Early Versus Standard Drainage Removal After Pancreatic Resections NCT00931554||Universita di Verona||Patients with low risk of PF, drain can be removed on POD3, prolong period of drain is associated with higher rates of PO complications with ncreased length of hospital stay and costs.|
|Comparative Study Between Duct to Mucosa and Invagination Pancreaticojejunostomy After Pancreaticoduodenectomy: (PJ) NCT02142517||Mansoura University||Invagination of PJ is not associated with reduction in PF rates|
|Randomized Controlled Trial on pancreatic Stent Tube in Pancreaticoduodenectomy NCT00628186||Wakayama Medical University||PF occurrence rate similar in both the procedures|
|39 Sandostatin in the Prevention of Postoperative Complications After Pancreaticoduodenectomy (PD)||Mansoura University||Ongoing|
|Trial Assessing Roux-en-Y Anastomosis of the Pancreatic Stump to Prevent Pancreatic Fistula Following Distal Pancreatectomy NCT01384617||Wakayama Medical University||Ongoing|
|Comparison of Feasibility Between Internal and External Pancreatic Drainage in Pancreaticoduodenectomy NCT01023594||Seoul National University Hospital||Ongoing|
The treatment offered by pancreatic resection has evolved through times and has proved to be highly effective yet surgically demanding procedure. The complications that arises in the form of PF adds significant morbidity and mortality, thus demands more sophisticated researches. Pancreatic surgeons have spent tremendous efforts in finding out effective strategies to reduce PF rates and sadly it hovers to around 15%.
Thus in light to its occurrence, a vigilance applied to study its important risk factors like, morphological structure, its texture, the status and dimensions of the common bile duct, the surgically critical sites of the tumor (ampullary, duodenal, cystic and islet pathology as well as the amount of blood loss; would surely curtail PF rates. Skills in advanced diagnostic methods in the form of CT and MRI would surely help in identifying these risk factors as well as early detection of complications chiefly PF. Surgically, duct to mucosal double layered anastomosis have been successful in reducing the PF rates and its validation is still awaited from the trials. Similarly, varieties of drainage system and pharmacological interventions have emerged in studies that are yet to be validated. The outcomes of trials are eagerly awaited and the target identified does offer us a strong hope in PF management.
Pancreatic fistula is an iatrogenic process and the resultant morbidity as well as mortality is its sequel. Risk to benefit ratio surpasses the severity of complications, yet a surgical induced and identifiable PF needs to be reduced to negligible rates, and we hope the future researches can deliver such which is still an unmet surgical need.
Authors have no conflict of interests to declare.