We perform reconstruction with the "omega loop". Dissection of the retropancreatic tissue is done along the right side of the upper mesenteric artery because it is known that circumferential dissection of the artery can lead to persistent, difficult to treat diarrhea and DGE due to loss of gastrointestinal coordination. Also we transect duodenum before ligament of Treitz. In this way, the ligament of Treitz and the initial jejunum remains in place and undamaged, allowing the nerves to remain intact retaining also the neurovascular integrity of the initial segment what we use in the reconstruction, because everything stays on its anatomical location.

In patients with a hard pancreas and a dilated pancreatic duct (diameter 3 mm) we perform duct-to-mucosa pancreaticojejunostomy. In patients with a soft pancreas and a nondilated duct we perform end-to-side invagination pancreaticojejunostomy with “U” shape sutures.

To my knowledge, not a few Japanese surgeons preferred pancreatogastrostomy to pancreatojejunosomy because of the simplicity and less occurrence of leakage in 1990s. However, most surgeons abandoned the pancreatogastrostomy because it is believed that patency of the anastomosis is not as good as pancreatojejunosomy, which resulted in poor functions in both of the pancreatic exocrine and endocrine glands. We actually sometimes encounter cases necessary for undergoing reanastomosis owing to the stricture of pancreatogastrostomy. I hope I can find articles on the issue, but at the moment there is no such papers.

This is largely a matter of personal preference with a particular technique. I am comfortable with pancreato jejunostomy: duct to mucosa variant for duct sizes > 2mm and the invaginating technique for ducts

I prefer pancretico-jejunostomy with duct to mucosa anastomosis. Invagination technique I did use, but I do not longer, because it can traumatize the pancreatic remnant.

Marko Zelic considerations are interesting and shareable. I make a duct to mucosa anastomosis if it is technically feasible (at least 4 "U" shape sutures) regardless of the diameter of the pancreatic duct.

Piero Chirletti SAPIENZA University of Rome

In all cases, I prefer the pancreaticojejunostomy with intussusception and stent in Wirsung. In particular, I adopted the reconstruction with two jejunal loops: the first loop at the stomach and the second loop at pancreas and biliry tree. For the anastomosis, I use tycron stitches 2 or 3/0 with pladget that provide for a better seal even in presence of the soft pancreas.

in the hands of an expert surgeon with a high volume, any technique will bring good results.. the problem comes when you are working in a low to medium volume center and there is no high volume center to refer the patient to...

one problem with randomized controlled trials of interventions is that the procedures are standardized and performed by experts.. yet we try to generalize the results to people who are neither that expert nor follow the techniques to the described standards in the trial..

working in a low to medium volume center for pancreatoduodenectomy where both PG and PJ is performed, my impression is to assess the texture of the pancreas and preferably perform PG if the pancreas is soft. i must admit that this is my own impression and constitutes the lowest possible level of evidence.. however, no high quality evidence is applicable to a situation other than high volume centers..

regards, raza

Even in a low to medium volume centre, one should continue to practice the anastomotic technique with which one is familiar and conversant.

I personally prefer duct to mucosa anastomosis, irrespective of the size of the duct and texture of the gland. In my experience I find significantly low leak rate with this technique when compared with single layer dunking that I used to practice earlier.

Thank you Dr Pal, I agree it comes down to what one is more comfortable with in one's experience. I also feel that in addition to the surgical technique, it is the experience with the postoperative management and available multidisciplinary support that improves the outcome of the patient.

regards, raza

I agree entirely.

Tobias:

I have recently reviewed this issue, so I will provide a condensed summary of my findings below:

POST-PANCREATICODUODENECTOMY (PD) RECONSTRUCTION: PANCREATOGASTROSTOMY (PG) V PANCREATICOJEJUNOSTOMY (PJ): A REVIEW

Confining my attention to Level I data (prospective randomized trials + meta-analyses and systematic reviews of RCT data), the landmark prospective randomized University of Verona trial [1] compared the results of pancreaticogastrostomy (PG) versus pancreaticojejunostomy (PJ) following pancreaticoduodenectomy in a prospective and randomized setting. Patients receiving PG showed a significantly lower rate of multiple surgical complications. PG was favored over PJ due to significant differences in postoperative collections, delayed gastric emptying, and biliary fistula. Thus it was concluded that although the the procedures did not show any significant differences in the overall postoperative complication rate, nonetheless biliary fistula, postoperative collections and delayed gastric emptying were all significantly reduced in patients treated by PG compared to PJ, and in addition, PG was associated with a significantly lower frequency of multiple surgical complications. This confirms the early findings of the prospective randomized Johns Hopkins Trial [2], and the Nihon University study [3].

The recent Hubei University of Chinese Medicine meta-analysis [4] of four RCTs found a significant difference in the morbidity of intra-abdominal complications (OR=0.34), supporting the conclusion that PG is better than PJ for pancreatic reconstruction after PD. Furthermore, in a just published Belgian multicenter randomized trial [5], in patients undergoing pancreaticoduodenectomy (PD) for pancreatic head or periampullary tumors, PG was found to be a more efficient procedure than PJ in the reduction of the incidence of postoperative pancreatic fistula (and I also note that more events in the PJ group were of grade ≥3a than in the PG group).

Finally, I anticipate that the in-progress German RECOPANC (RECOnstruction after partial PANCreatoduodenectomy) trial [6,7] comparing the two procedures, PG and PJ, across 14 centers for pancreatic surgery should, I predict, provide when it reports further decisive confirmation of the more optimal benefits of pancreaticogastrostomy (PG) over pancreaticojejunostomy (PJ) for post-pancreatoduodenectomy (PD) reconstruction.

METHODOLOGY OF THIS REVIEW

A search of the PUBMED, Cochrane Register of Controlled Trials, MEDLINE, EMBASE, AMED, CINAHL, PsycINFO, (WoS) Web of Science, BIOSIS, LILACS and Scirus databases was conducted without language or date restrictions, and updated again current as of date of publication, with systematic reviews and meta-analyses extracted separately. Search was expanded in parallel to include just-in-time (JIT) medical feed sources as returned from Terkko (provided by the National Library of Health Sciences - Terkko at the University of Helsinki). A further "broad-spectrum" science search using Scirus (410+ million entry database) was then deployed for resources not otherwise included. Unpublished studies were located via contextual search, and relevant dissertations were located via NTLTD (Networked Digital Library of Theses and Dissertations) and OpenThesis. Sources in languages foreign to this reviewer were translated by language translation software.

REFERENCES

1. Bassi C, Falconi M, Molinari E, et al. Reconstruction by pancreaticojejunostomy versus pancreaticogastrostomy following pancreatectomy: results of a comparative study. Ann Surg 2005; 242(6):767-71, discussion 771-3. At: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1409871/.

2. Yeo C, Cameron J, Maher M, et al. A prospective randomized trial of pancreaticogastrostomy versus pancreaticojejunostomy after pancreaticoduodenectomy. Ann Surg. 1995;222:580–588. At: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1234894/.

3. Takano S, Ito Y, Watanabe Y, Yokoyama T, Kubota N, Iwai S. Pancreaticojejunostomy versus pancreaticogastrostomy in reconstruction following pancreaticoduodenectomy. Br J Surg. 2000 Apr;87(4):423-7. At: http://onlinelibrary.wiley.com/doi/10.1046/j.1365-2168.2000.01395.x/abstract;jsessionid=7CE566F2F1150AE1C8C08A1D18B9C585.d02t01.

4. Shen Y, Jin W. Reconstruction by Pancreaticogastrostomy versus Pancreaticojejunostomy following Pancreaticoduodenectomy: A Meta-Analysis of Randomized Controlled Trials. Gastroenterol Res Pract 2012; 2012:627095. At: http://www.hindawi.com/journals/grp/2012/627095/.

5. Topal B, Fieuws S, … Belgian Section of Hepatobiliary and Pancreatic Surgery. Pancreaticojejunostomy versus pancreaticogastrostomy reconstruction after pancreaticoduodenectomy for pancreatic or periampullary tumours: a multicentre randomised trial. Lancet Oncol 2013; 14(7):655-62. At: http://www.thelancet.com/journals/lanonc/article/PIIS1470-2045%2813%2970126-8/fulltext.

6. Pancreatogastrostomy versus pancreatojejunostomy for RECOnstruction after partial PANCreatoduodenectomy – A randomized controlled trial. Deutsches Register Klinischer Studien/DRKS (German Clinical Trials Register). DRKS-ID: DRKS00000767. At: https://drks-neu.uniklinik-freiburg.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00000767.

7. Wellner UF, Brett S, Bruckner T, et al. Pancreatogastrostomy versus pancreatojejunostomy for RECOnstruction after partial PANCreatoduodenectomy (RECOPANC): study protocol of a randomized controlled trial UTN U1111-1117-9588. Trials. 2012 Apr 27;13:45. At: http://www.trialsjournal.com/content/13/1/45].

This is a matter of one's personal preference and experience. However, the pancreaticogastrostomy seems to be the method of choice. On another hand, the PJ may produce a septic reaction, necrosis of the pancreatic stump, leakage more frequently, etc. The stomach lumen is almost sterile, and that of the small intestine is not.

I prefer duct to mucosa pacreatico-jejunostomy. Because it is more secuer and physiologic and faster than pancraticogasterostomy.In my experience ( data is not yet poblished) panceratic fistula incidence is lower than other type of reconstruction.

May be Tobias will share some interim data from the RECOPANC trial?

Recently we have changed our technique for the soft pancreas from a duct-to-mucosa PJ to a pancreaticogastrostomy with very good results.

I agree with the statement that the best technique is what you are comfortable with, irrespective of the trials. I am comfortable doing PJ double layer with duct to mucosa, with internal /external stenting. I have seen people performing PG, but have never performed one. I do not wish to change despite the reports stating it has lower leak rates etc. Stenting is again controversial, I have changed twice: external to internal to external again, matter of comfort.

In the beging of my training i used to do PJ anastomosis but unforuntaly the pancreatic leakage and fisula was very high ,with more experience i shifted to PG technique with good outcome and confertable results.

intersting discussion, I still do the PJ, probably this is the common reconstruction in the UK, my leak rate less than 10%, however, if there will be level I evedence to support PG, I will train myself to do it.

I think a 3 anastomosis technique is better, pancreatico jejunostomy(ducto mucosal) then hepatico jujenostomy(ducto mucosal) and at last gastro jejunostomy (end to side or side to side)

PJ is easier to do, faster and doe not require extensive mobilization of the pancreatic stump.

As Gastroenterologist, I would like to focus on exocrine pancreatic function that is better preserved by PJ than PG (preliminary data to be published).

During my stay at Hannover Medical School I used to perfom PG, then I moved to the University of Essen where we perormed PJ, so I have the experience with both techniques. Now, I prefere PJ and use PG only in complications after PJ. However, for me there is no definitive answer to this question. Both techniques are of equal value.

I'm comfortable doing PJ single layer with internal stenting. I prefere PG only in soft pancreas and lack/small pancreatic duct.

Studies have shown no difference in the leak incidence between the two techniques. The preferred technique is what you think you do well.

I usualy perform a PJ. Only if i am dealing with pancreatosteatosys or IPMN i perform a PG.

I am performing a PG by IPMN so that the gastroenterologist can perform a gastroscopy and check the pancreas by taking samples. That's the only reason. In my experience I never had an insufficiency due to fibrosis after an PG

We have examined the rate of exocrine and endocrine insufficiency after PG and PJ. Unfortunately after either form of reconstruction the rate of clinical exocrine insufficiency after 2 years is over 50%. Meaning that at least half of the patients are depending on enzyme supplementation. Between PG and PJ there was no difference in this retrospective observation.

There is in addition a theoretical benefit of pancreatogastrostomy which remains open to be confirmed by experimental research. Pancreatic enzyme precursors as trypsinogen are activated by cathepsin B or enterokinase. In the case of a PJ enterokinase is abundantly present in case of a small leakage. I would be interested if anybody has some data on this?

Did anyone investigate those patients before operaton to establish their stomach condition, namely, acidity, presence of atrophy, colony forming units count of its content, etc., and the small intestine bacterial overgrowth syndrome presence, so that one's decision to be based not only on personal preferencies or/and anatomical traits?

In Girona (Catalonia) we prefer PG over PJ.We recently published our experience (Br J Surg. 2013 Nov;100(12):1597-605)

I feel that the question must be answered in each center as far as even when you say PG or PJ probably, from a technical point of view, we are talking about different procedures

This is one of my latest PPPD with PJ anastomoses single layer with 4-0 monocryl. Does anyone else have fotos to show us his technic?

Fj

Hepatojejunostomy

Hepatojejunostomy

Gastrojejunostomy single continous layer 3-0 monocryl.

I present a simple and dependable technique for

pancreatico-enteric anastomosis with a soft pancreas and non-dilated duct (≤3mm). I use for PJ a polydiaxone 5-0 (PDS, 5-0)

Fig. 1. The pancreatic remnant is assessed for texture (soft pancreas) and the size of the main pancreatic duct (diameter < 3mm).

Fig.2. Posterior ductal sutures.

Fig.3. Anterior and posterior ductal sutures.

Fig.4. Posterior layer - horizontal mattress sutures.

Fig.5. Posterior wall of the pancreatic duct is now anastomosed to posterior wall of the opened jejunum.

Fig.6. The anterior wall of the pancreatic duct is now anastomosed to the opened jejunum anteriorly.

Fig.7.Anterior layer - interrupt sutures.

Step one: Posterior ductal sutures

Step two: Anterior ductal sutures

Step three: Posterior layer - horizontal mattress sutures.

Step four: Posterior wall of the pancreatic duct is now anastomosed to posterior wall of the opened jejunum.

Step fiive:The anterior wall of the pancreatic duct is now anastomosed to the opened jejunum anteriorly.

Anterior layer - interrupt sutures.

It is really a very nice and easy anastomoses. When i am dealing with soft pancreas i prefer a PG anastomoses. Do you leave the tube in place or do you take it out when you perform the hepatojejunal anastomosis?

The stent in the pancreatic duct is type "Perdue".

Dr.Kostov, thank you for the nice presentation and the highest quality pictures!

Can you share your experience with us as to some special methods or tricks you usually perform so that to avoid the jejuno-pancreatic reflux of the chyme through that U-shape loop, save the (Braun) side-to-side additional anastomosis, if do that at all? And what has been the longest time for a 'forgetting' stent to stay in place in your practise?

I’m using this technique for PJ, over a 3-year period. The pancreatico-duodenectomy is pylorus-preserving without Braun. PJ is isoperistaltic, which is a prerequisite for a lack of jejuno-pancreatic reflux. Practically there is no clinical and biochemical evidences for a reflux. The longest time for a “forgetting” stent is 2 months.

Thank you. On my experience, a high fever was characteristic for the first several days after the PJ, that made me to abandone that method. I think so that the fever occured due to the jejuno-pancreatic reflux.

We do not perform the pylorus-preserving pancreato-duodenectomy because the only indication in my university clinic for the Whipple is the cancer of the papilla or that one of the head, so that the pylorus salvation would be not a radical procedure.

pancreato jejunostomy (after work pancreatoduodenectomy! without preerving the pylore, half removal of the pancreas etc)....Dr marchal (clinique du millenaire, montpellier ) did a great job:-) moreover, no need of pnacreatic enzyme supplementation:-)

No definitive study showed superiority of a technique over another, but the choice is left to each surgeon preference. However, PG seems to be less challenging in the hands of less experienced surgeons. Some of the complications (i.e bleeding) could be treated endoscopically

We perform pancreatico-jejunostomy with external stent inserted across the anastomosis to drain the pancreatic duct. This has led to a dramatic decrease of both pancreatic fistula and morbidity rates.

I prefer Traverso-Longmire reconstruction in all cases where it is possible.

Unfortunately I collected a couple of severe bleeding after pancreaticogastrostomy, but I did not abandoned that kind of reconstruction. Probably i missed some tricks and tips of the technical approach.

Our experience is based on over 140 cases of duodenopancreatectomy, in which we have always carried out a reconstruction with double loop. The first jejunal loop is used for the anastomosis with the stomach, which is stored above the pylorus; the second loop, about 30 cm from the first one, is used for the pancreatic and biliary anastomosis. We then perform a pancreatic jejunostomy end-to-end, with invagination of the pancreatic stump after application of a catheter into the Wirsung.

At the end, the pancreatic anastomosis is coated with a patch of Tachosil. The overall incidence of pancreatic fistula, in the last 100 cases, only grade A and B, has been 4.2% of cases. Mortality 1%.

The choice of using the first jejunal loop for the gastric anastomosis is based on the pathophysiological concept that in the mucosa of the first jejunal loop are present many of entero-hormones receptors that, if activated, by transit of bolus, allow to maintain high the exocrine secretion of pancreas, with reduction of malabsorption.

I prefere PJ

PJ is the standard of care in pancreas reconstruction after PD.

No advantages in terms of morbidity are shown using PG and, theoretically, draining the pancreas in the stomach, you can lose the exocrine function of the gland.

One of my last whipple operations is a 64 year old male where i performed a PJ anastomosys.On post op day 5 i have a combined PF and bile leak of 100cc per day without any clinical impact on the patient.He is taking oral food no fever no haemorrage.How would you manage this patient? The AMS level from the drain is 100000 and the bile 12.

we have been practicing both PJ and PG.I am not sure in less experienced, PG is simpler,both require equal competence.PG is associated with bleeding from the pancreas,if it leaks ,the fistula is from pancreas,stomach and small bowel.Our preference is side to side PJ with internal stenting and two layer tecnique.I appreciate the ref given by CK.

It has been over six months since my last review of this issue (8 Sept 2013), a summary of whose findings I posted above, and so I will here offer an update of newly reported evidence, and a new appreciation of where we stand in this debate.

SUMMARY OF DATA UP TO FIRST REVIEW

As already noted in my previous posting above, the preponderance of systematically reviewed, critically appraised and  methodologically assessed data support the higher safety of pancreaticogastrostomy (PG) versus pancreaticojejunostomy (PJ). Thus, despite an occasional and sporadic negative trial [2], one study [3] (n=142) found that the incidence of pancreatic fistula (PF) in the PG group (0%) was significantly less than that after PJ (13%), while rates of Intra-abdominal hemorrhage (4%) and intra-abdominal abscess (6 per cent) in the PJ group contrasted to zero after PG, and furthermore with two hospital deaths (3 per cent) in the PJ group, none in PG, while another study [1] found that patients receiving PG showed a significantly lower rate of multiple surgical complications, with PG being favored over PJ due to significant differences in postoperative collections, delayed gastric emptying, and biliary fistula, and still again, the  large Belgian Lancet RCT [5] (n=329) found that (1)  more events in the pancreaticojejunostomy group were of grade ≥3a than in the pancreaticogastrostomy group, and (2) pancreaticogastrostomy was more efficient than pancreaticojejunostomy in reducing the incidence of postoperative pancreatic fistula, and yet again the Shen meta-analysis of 4 RCTs [4] (n=553) conclude that (1) significantly greater rate of  morbidity of intra-abdominal complications in PJ compared to PG, and (2) PG is superior to PJ for pancreatic reconstruction after PD.

UPDATE: NEW STUDIES SINCE LAST UPDATE (Since Sept. 8, 2013)

In a meta-analysis of five RCTs [8] it was found that: (1) patients showed decreased incidence of fluid collection (FC) with PG, and (2) in four of the five RCTs (n=553) PG patients both showed decreased incidence of overall morbidity and biliary leak.  And another recent RCT [10] (n=123) found that the incidence of pancreatic fistula was significantly higher following PJ than for PG, as was the severity of pancreatic fistula, and the hospital readmission rate for complications, as well as weight loss, was significantly lower after PG, along with better exocrine function.

Another recent meta-analysis of four randomized controlled trials (RCTs) and 22 observational clinical studies [9] found that PG was associated with lower rates of pancreatic fistula (PF) and intra-abdominal fluid collection, including intraperitoneal abscess and asymptomatic effusion, as compared to PJ, the difference being statistically significant.

In addition, the just published massive French Surgical Association multicenter study [11] of 1325 patients undergoing a PD for ductal adenocarcinoma at 37 institutions found that  PF occurred more frequently after PJ compared with PG, with clinically relevant PF (grade B and C) as well as  severe complications (Dindo–Clavien grade IIIB, IV, V) also significantly more frequent after PJ than PG, concluding therefore a significantly higher incidence and clinical severity of PF being associated with PJ (that is, higher rate of major post-operative morbidity compared with PG). In addition, and importantly, PF remained the leading cause of lethal complications, with the majority of deaths directly attributable to PF associated with intra-abdominal, intestinal bleeding and sepsis.

And as I previously noted, there have been four prospective randomized trials Over the past 20 years comparing PJ and PG [16-19], but while three of these studies [16-18] have reported no difference in the pancreatic fistula (PF) rate, these were of lower methodological quality and were rated Grade B [9], with only one RCT [19] rated Grade A methodological quality of the evidence [9], that one demonstrating a significantly lower rate of PF after PG than PJ (4% vs. 18%).  And a meta-analysis comparing PG and PJ [12] confirmed a significant decrease in the incidence of PF and mortality for PG.

Note also that data [13] has reported a shorter and more favorable evolution in PF after PG compared with PJ, with another study [14] reporting a significant reduction in relaparotomy and completion pancreatectomy rates with PG over PJ (with conservative management of PF after PG was successful in more than 75% of patients [15]), and that in general, PF after PJ increased the use of interventional or operative management compared with PJ [13,14].

WHAT THE BALANCE OF THE EVIDENCE SAYS: LESSONS LEARNED

Therefore, new and updated systematically reviewed, critically appraised and  methodologically assessed data since last review (8 Sept 2013) further confirms the more optimal benefits and safety of pancreaticogastrostomy (PG) over pancreaticojejunostomy (PJ) for post-pancreatoduodenectomy (PD)  reconstruction. This is not to say that in the hands of a gifted surgeon with exceptional expertise in execution of pancreaticojejunostomy (PJ) procedures, highly positive outcomes cannot be delivered; but in the general case of the execution of  pancreaticojejunostomy (PJ) procedures compared to pancreaticogastrostomy (PG) across all surgical modalities and contexts, the weight of the best evidence to date favors PG along a number of safety, complications, morbidity and mortality parameters of comparison identified above, and I would again predict from my own sources that the in-progress German RECOPANC  trial [6,7] comparing the two procedures across 14 high-volume centers for pancreatic surgery is likely to provide further decisive confirmation - if any is needed - of the more optimal broad-spectrum safety of pancreaticogastrostomy (PG) over pancreaticojejunostomy (PJ) for post-pancreatoduodenectomy (PD)  reconstruction.  

METHODOLOGY OF THE REVIEW

A search of the PUBMED, Cochrane Library / Cochrane Register of Controlled Trials, MEDLINE, EMBASE, AMED (Allied and Complimentary Medicine Database), CINAHL (Cumulative Index to Nursing and Allied Health Literature), PsycINFO, ISI Web of Science (WoS), BIOSIS, LILACS (Latin American and Caribbean Health Sciences Literature), ASSIA (Applied Social Sciences Index and Abstracts), and SCEH (NHS Evidence Specialist Collection for Ethnicity and Health) was conducted without language or date restrictions, and updated again current as of date of publication, with systematic reviews and meta-analyses extracted separately. Search was expanded in parallel to include just-in-time (JIT) medical feed sources as returned from Terkko (provided by the National Library of Health Sciences - Terkko at the University of Helsinki). Unpublished studies were located via contextual search, and relevant dissertations were located via NTLTD (Networked Digital Library of Theses and Dissertations) and OpenThesis. Sources in languages foreign to this reviewer were translated by language translation software.

REFERENCES

Physiologically, I prefer PJ. There are various types of PJ anastomosis. You can not compare one PJ and PG and then make the conclusion. It is not logical. My personal opinion is that a perfact PJ is still not standardized yet. I use biliary diversion to prevent bile activation of the pancreatic enzymes in case there is a chance of PJ leakage and develope whole thickness pancreas suture technique for end to side PJ to eliminate PJ leakage. I find this technique is very simple and easy, no matter the small duct size or soft pancreatic texture. No internal stent is necessary.

We have examined the rate of exocrine insufficiency after PG and PJ  after 1 year  too. The rate of insufficiency is over 70% and was difference between PG and PJ.

I prefere PJ for several reasons: it maintain the exocrine function, is at the moment the gold standard, but mostly, there are not advantages with the use of PG (no prospective randomized trials showed any advantages). However, for high risk pancreas, the Finnish "no-touch" anastomosis can be a very interesting alternative to the conventional PJ. The published results are very encouraging 

leaks are found in both techniques,I feel the technique which is good in your hands is the best.If a PJ leaks it is mostly a pancreatic fistula where as if PG leaks it is pancreatic,gasric and biliary(duodenum) fistula.