Saudi Journal of Gastroenterology
Home About us Instructions Submission Subscribe Advertise Contact Login    Print this page  Email this page Small font sizeDefault font sizeIncrease font size 
Users Online: 928 

ORIGINAL ARTICLE Table of Contents   
Year : 2007  |  Volume : 13  |  Issue : 2  |  Page : 70-75
Longitudinal pancreatico-gastrostomy: An effective means of pain control in chronic pancreatitis

Dept. of Surgery, I.P.G.M.E.R, S.S.K.M Hospital, Kolkata - 700020, West Bengal, India

Click here for correspondence address and email


Aim: To assess the effectiveness of longitudinal pancreatico-gastrostomy (LPG) in relieving intractable abdominal pain in patients of chronic pancreatitis (CP) with dilated main pancreatic duct (MPD). Materials and Methods: This prospective study was conducted from 1997-2005 at two university-affiliated hospitals in India. Ductal decompression by LPG was performed in 30 patients (26 males, 4 females) suffering from intractable pain due to CP. The operative outcomes were classified as satisfactory and unsatisfactory according to whether the patients were completely / almost completely relieved of pain or continued to be troubled by pain. Main outcome measures: The main outcome measures were pain-relief, postoperative morbidity, and mortality. Results: There was no postoperative mortality. Morbidity included pancreatic fistula in one patient, which closed spontaneously, gastric hemorrhage in two, prolonged ileus in one and wound infection in another. Twenty eight patients (93%) patients could be followed up till the end, the mean follow-up period being 23.8 (range 3-96) months. Operative result was satisfactory (no pain or mild pain) in 23 (82%) and unsatisfactory (moderate to severe pain) in 5 (18%) patients. Complete pain relief was seen in 14 (50%) patients. The functional results, in terms of pain relief, were better in patients who had abstained from alcohol postoperatively. Conclusion: LPG is a good operative procedure to relieve intractable pain in patients of CP with an MPD diameter of at least 7 mm.

Keywords: Chronic pancreatitis, pancreatico-gastrostomy, pancreatico-jejunostomy, surgical management

How to cite this article:
Bhattacharjee PK. Longitudinal pancreatico-gastrostomy: An effective means of pain control in chronic pancreatitis. Saudi J Gastroenterol 2007;13:70-5

How to cite this URL:
Bhattacharjee PK. Longitudinal pancreatico-gastrostomy: An effective means of pain control in chronic pancreatitis. Saudi J Gastroenterol [serial online] 2007 [cited 2022 Jul 3];13:70-5. Available from:

Chronic pancreatitis (CP) is a challenging and frustrating clinical problem. Surgical treatment of CP is directed mainly towards the relief of intractable pain. Majority of the patients have already undergone a long period of unrewarding conservative treatment and are most likely to be those with the worst prognosis. Operative treatment falls into three categories: resection of the gland, ductal drainage and denervation (celiac ganglionectomy). The choice of the procedure is based on the main pancreatic duct (MPD) morphology. Ductal drainage is useful in dilated and obstructed ducts while pancreatic resection is suitable for small duct disease. [1]

Though pancreatic resection, especially near total (80-95%) pancreatectomy, is most effective in relieving pain, it is associated with unacceptably high postoperative morbidity, mortality, exocrine and endocrine insufficiency. [1] Overall, best results are reported with ductal drainage procedures performed on patients with dilated MPD. [2] This study was undertaken to evaluate the effectiveness of longitudinal pancreatico-gastrostomy (LPG) in achieving pain control in patients of CP suffering from intractable abdominal pain.

   Materials and Methods Top

This prospective study conducted over a period of 8 years between May 1997 and April 2005 at two university-affiliated hospitals in India, included 30 patients of chronic pancreatitis (26 males and 4 females). The median age of the patients in this series was 34 (range 17-59) years; 17% being over the age of 40 years. Mean duration of illness was 3.35 (range 2-5) years. Median preoperative body weight was 55 (range 38-68) kg. Weight loss was present in 21 (70%) and diabetes in 11 (36.7%), six of whom were insulin-dependent.

The inclusion criteria were CP associated with severe, recurrent and intractable abdominal pain with an MPD diameter of ≥7 mm. Pain severity as assessed both preoperatively and postoperatively was categorized as none to mild if patients had no pain or pain which did not interfere with everyday life and required nothing stronger than occasional outpatient medication. Moderate pain was defined as intermittent or constant pain severe enough to require the same oral analgesics frequently. Severe pain requiring even hospitalization, warranted administration of more potent narcotics because of its unbearable, intermittent or constant nature.

Though tools such as the visual analogue scale are very useful for pain assessment, we had opted for the relatively simpler means of evaluating pain verbally as it required the patients to possess minimal verbal and language recognition skills only. It was helpful while dealing with some of the patients with average or below average intelligence, without reading and / or writing skills and with those posing language barriers because of their different ethnic backgrounds.

Disease-related exclusion criteria were CP in patients with small duct disease or MPD < 7 mm in diameter. Patient-related exclusion criteria were gross cardiovascular abnormality, history of myocardial infarction within the last 6 months, detection of pancreatic malignancy or coexisting malignancy of other organs. Appropriate informed consent was obtained from all patients and the ethics committees of both the study centers approved the study protocol. Diagnosis of CP was made from suggestive history, clinical examinations and routine diagnostic workup like skiagrams of the abdomen, ultrasonography (USG), Computed tomography (CT) scans of the abdomen and endoscopic retrograde cholangio-pancreatogram (ERCP).

The skiagram was useful in detecting stones in eight out of ten cases where it was advised. USG was done in all 30 cases and gave positive information in all. ERCP was performed in ten and gave positive information [Figure - 1] in 90% (n=9), while cannulation failed in one patient. Out of the five patients with gallstone-related pancreatitis, three were detected to have small common bile duct (CBD) stones as well and had them removed endoscopically while the other two were detected to be having papillary stenosis and underwent endoscopic sphincterotomy. No attempts were made for extraction of any pancreatic stones endoscopically. CT scan was done in 20 patients and were positive in 95% (n=19). The median MPD diameter was 9 (range 7-15) mm.

All the selected patients had two or more of the following three internationally accepted criteria for CP: pancreatic calcification in the skiagram of the abdomen, decreased pancreatic exocrine functions (judged clinically if diarrhea or steatorrhoea was treatable by pancreatic enzyme substitution and by fecal fat more than 6 g in 24 h) or abnormality of the MPD (as judged by dilated MPD, obstruction in the MPD, stones in the duct or small pancreatic pseudocyst in USG or ERCP). [3] Fecal fat estimation was advised in 6 (20%) patients presenting with features of steatorrhoea and all had results of > 6 g in their 24 h fecal samples and needed enzyme supplementation.

CP, in this series, was attributable to chronic alcohol ingestion (n=14), juvenile tropical pancreatitis (n=6), biliary stone disease (n=5), trauma (n=1) and idiopathic (n=4 patients). Juvenile tropical pancreatitis is characterized by exocrine and endocrine pancreatic failure, pancreatic calcification, abdominal pain mostly in nonalcoholic patients < 25 years of age. [4] The preoperative patient details are summarized in [Table - 1].

To avoid bias, the choice of the patients on the basis of the inclusion criteria and the assessment of the results were done by different individuals, none of whom participated in the study. All patients underwent ductal decompression by LPG and were operated on by the author though the other members of the team varied.

Operative procedure

All ductal decompressions of this series were primary procedures. The pancreas was approached via the gastrocolic omentum. The dilated MPD was localized by palpation of the anterior surface and by needle aspiration of clear pancreatic juice. The duct was completely laid open from the tail end of the gland right up into the neck and head, all strictures being opened up in the process. Towards the duodenal wall, the incision was kept within about a centimeter of the duodenum to avoid injury to the pancreatico-duodenal vessels. Probing the duct helped in locating the distal-most stricture. A piece of tissue from the margin of the opened duct was sent for histopathological examination. Intraductal calculi were removed as far as possible from the entire length of the duct [Figure - 2].

Anastomosis was then performed with the posterior surface of the stomach in two layers: the outer between the pancreatic capsule and the seromuscular layer of the stomach with (2-0) black silk and the inner between the pancreatic duct and the full thickness of the stomach with (2-0) polyglactin (Vicryl). The stomach was opened up in short segments during the first through-and-through layer of suture to avoid excessive bleeding from the cut stomach wall [Figure - 3]. Cholecystectomy was performed routinely.

All patients received 100 g octreotide (an octapeptide analogue of somatostatin which inhibits the basal and stimulated exocrine pancreatic secretion) subcutaneously every 8 h for the first seven postoperative days. Once oral intake was normalized, all patients received high lipase-containing, enteric-coated pancreatic enzymes and oral citrate preparations (a known solvent of pancreatic stones).

The primary endpoint of the study was return to normal activities. The main outcome measure was pain relief, mortality and morbidity rates of the procedure. As the emphasis was principally on the main stated outcome measure, i.e., achievement of pain control and low morbidity, other validated quality of life (QOL) questionnaires were not utilized during follow-up.

Postoperative evaluation was done, initially on a monthly basis for 6 months and yearly thereafter. Relief of pain was assessed by comparing it with its preoperative severity. When evaluating operative outcome in terms of pain relief after the operation, patients who were completely relieved of pain or had mild, tolerable pain which did not interfere with everyday activities, were grouped as satisfactory. Patients still suffering from episodic moderate pain requiring analgesics or persistent severe pain requiring regular analgesics / opioids or hospitalization were categorized as unsatisfactory.

   Results Top

At the time of presentation, none of the patients had jaundice. Blood biochemistry suggested hypoalbuminemia in 12 (40%), mildly elevated serum amylase in 4 (13.3%) patients and elevated alkaline phosphatase in 5 (16.7%) patients. The mean operating time was 83 (range 75-90) minutes. No blood transfusion was required perioperatively. Presence of pancreatic stones / calcification was detected in 23 (76.7%) patients. Histopathological examination confirmed all patients to be suffering from CP. There were no postoperative deaths. Postoperative complications were noted in 16.7% (n=5 patients), all of which could be managed conservatively. One (3.3%) patient developed a pancreatic fistula that closed spontaneously after 20 days of supportive management and maintenance of operatively-placed drains. Pancreatic fistula was confirmed by the presence of amylase-rich fluid (> 1000 U/L) in the drain. Two (6.6%) had gastric hemorrhage in the early postoperative period. Bleeding was controlled after treatment with hemostatics (tranexamic acid), ice-cold lavage and blood transfusions. One patient had prolonged ileus and another had a wound infection. Median period of hospitalization was 8 (range 7-20) days.

One patient was lost to follow-up after 11 months while another died 18 months after the operation due to complications of continued alcohol misuse. Long-term results were assessed by compilation of results from the 28 remaining patients who could be followed up till the end. Three patients continued to drink postoperatively inspite of advice on the contrary. Mean period of follow up was 23.8 (range 3-96) months. Pain relief was documented in 23 (82%) patients, out of which 14 (50%) patients had excellent results and were completely relieved of the pain. Nine (32%) patients experienced occasional pain of mild nature and were satisfied with tranquilizers. The functional results, in terms of pain relief, were better in patients who had stopped drinking alcohol. All 3 (11%) individuals who defied advice regarding abstinence from alcohol continued to experience severe pain and were dependent on opioids for relief. Another 2 (7%) individuals who had gross calcinosis of the pancreas continued to experience a moderate degree of pain during follow-up. Comparative pre- and postoperative patient data are summarized in [Table - 2].

Thus, the operative results in terms of pain relief were satisfactory (no pain or mild pain) in 23 (82%) and unsatisfactory (moderate to severe pain) in 5 (18%) patients. Amongst the patients below the age of 40 years, 68% (n=17) had satisfactory pain relief, while 60% (n=3) of those above 40 years of age continued to be troubled by pain. None of these patients with unsatisfactory results agreed to any further operative intervention for relief of pain. A total of 19 (67.9%) patients could return to their normal activities within a median interval of 7.5 months after operation. During the follow-up of the 28 patients, the number with diabetes increased from 11 (36.7%) to 16 (57.1%). However, diarrhea and steatorrhoea were mostly controlled because of substitution of high lipase-containing, enteric-coated preparations postoperatively.

   Discussion Top

As was true in our series, chronic alcohol consumption is the most common cause of chronic pancreatitis worldwide. [5] Pain is the most common presentation while its intractability and interference with the lifestyle of the individual is the primary indication of surgical intervention. [5] Surgery neither stops nor reverses the disease process-it is at best a palliative intervention aimed at slowing down the ongoing effects of chronic inflammation on the gland. [6] Abstinence from alcohol is probably the single greatest contributory factor to the positive result of surgery. [5],[7]

Though the value of drainage operations for pain in CP has never been proven in randomized controlled settings comparing drainage with sham operations, the results of individual operations suggest that ductal drainage in patients with dilated MPD offers the best operative results in CP. [3],[8] The ideal operation for patients with CP should achieve permanent pain relief without exacerbating pancreatic exocrine and endocrine insufficiency while having a low operative morbidity and mortality. At the same time, it should be easy to perform and independent of patient abstinence from alcohol. Tripodi and Sherwin in 1934 were the first to describe PG as a ductal drainage procedure following partial pancreatic resection. [9] Recently, some uncontrolled studies indicated a revival of interest in pancreatico-gastrostomy as a drainage procedure in chronic alcoholic pancreatitis though the techniques of pancreatico-gastric anastomosis are not always similar. [2],[3],[7],[10],[11],[12],[13],[14] Results with this procedure in a centre in southern India were reported to be excellent. [10] One controlled series comparing pancreatico-gastrostomy with coeliac blockade found better pain-relieving effect due to the PG operation after six months. [11]

A reliable comparison between pancreatico-gastrostomy and pancreatico-jejunostomy, per se, from published data, is difficult because the reported series are seldom related to chronic pancreatitis patients alone, usually reporting mixed data concerning mostly periampullary carcinoma. [15] There are some definite advantages of an LPG over pancreatico-jejunostomy. Technically the operation is easier to perform than pancreatico-jejunostomy and also easily undone if at a later date a reoperation becomes necessary. The stomach being in close proximity to the pancreas and with a thicker and a more vascular wall, holds sutures well and also enables nasogastric aspiration on contact with the anastomosis. It has the theoretical advantage of bathing the pancreatic duct in acidic secretion, thereby having a trypsin neutralizing effect. These factors reduce the chances of anastomotic leak and stenosis and also prevent a flare-up of pancreatitis in the immediate postoperative period. Moreover, PG enables leftover pancreatic stones, if any, to come in direct contact with its solvents namely the gastric acid and orally administered citrate. [10]

Unlike Roux-en-Y pancreatico-jejunostomy, it obviates interrupting small gut continuity and by reducing one anastomosis, it not only reduces the relative chance of anastomotic leak but also makes the operation relatively quicker. Furthermore, it offers a technically easier approach to verify the pancreatic duct permeability by endoscopic examination. [16]

LPG was found to be a safe procedure. There were no deaths in our series in the early postoperative period. Only one patient died 18 months after operation due to complications unrelated to the surgery. Many other authors have reported low mortality (range 0-4.4%) [2],[7],[11],[12],[13],[14] and the figures compare favorably with pancreatico-jejunostomy as a drainage procedure. [17],[18],[19] Morbidity of this procedure was low. Pancreatic fistula developed in only one patient (3.3%) early in this series. Though the incidence of pancreatic fistula was more than that reported in some other series with PG (where no incidence of fistula was reported), [2],[7],[14] it is certainly less than its reported incidence following pancreatico-jejunostomy (up to 5%). [19] Hemorrhage from the anastomotic site is another complication which has been reported by others [2],[7] and occurred in 6.6% of our patients. However, unlike others, we could manage it conservatively.

The safety of LPG is evidenced by unequivocal findings of low operative morbidity and absence of early mortality in our patients. This also accounts for the median duration of a hospital stay of only 8 days. We had attempted to make the stoma as wide as possible and avoided doing it in patients with MPD < 7 mm since it is documented that pancreatico-jejunostomy in those with stoma of > 6 cm [17] and MPD > 7 mm [19] experienced significantly better pain relief. The definition of a successful outcome of surgery for pain relief is difficult. We equated it with satisfactory pain relief, i.e., no pain or mild pain. This was the case in 23 (82%) of the 28 patients followed up over a mean period of 23.8 months. Other authors have reported similar experiences. [2],[3],[7],[11],[13],[14] Positive correlations between poor results in terms of pain relief and continued postoperative alcohol abuse supports the view that abstinence from alcohol is an important predictor of pain relief. [2],[3],[7],[13],[20],[21] This technique of pancreatico-gastrostomy should be further studied by randomized, controlled trials against other standard operations with assessment of functional results in terms of progression of exocrine and endocrine functions and weight gain. In conclusion, LPG is a simple and safe procedure of ductal decompression which gives effective relief from intractable pain in selected patients of chronic pancreatitis and merits a more widespread application.

   References Top

1.Marrow CE, Cohen JI, Sutherland DR, Najarian JS. Chronic pancreatitis: Long term surgical results of pancreatic duct drainage, pancreatic resection and near total pancreatectomy and islet cell autotransplantation. Surgery 1984;96:608-15.  Back to cited text no. 1    
2.Kovacs I, Arkossy P, Mahunka M, Sapy P. Gastric acidity following longitudinal pancreaticogastrostomy. Hepatogastroenterology 1998;45:895-9.   Back to cited text no. 2  [PUBMED]  
3.Ebbehψj N, Christensen E, Madsen P. Prediction of outcome of pancreaticogastrostomy for pain in chronic pancreatitis. Scand J Gastroenterol 1987;22:337-42.  Back to cited text no. 3    
4.Mabogunje OA, Lawrie JH. Surgery for chronic pancreatitis in Zaria, Nigeria. World J Surg 1990;14:45-7.  Back to cited text no. 4    
5.Cooperman AM. Surgery and chronic pancreatitis. Surg Clin North Am 2001;81:431-55.  Back to cited text no. 5    
6.Sakorafas GH, Farnell MB, Nagorney DM, Sarr MG. Surgical management of chronic pancreatitis at the Mayo Clinic. Surg Clin North Am 2001;81:457-65.  Back to cited text no. 6    
7.Pain JA, Knight MJ. Pancreaticogastrostomy: The preferred operation for pain relief in chronic pancreatitis. Br J Surg 1988;75:220-2.   Back to cited text no. 7    
8.Brinton MH, Pellegrini CA, Stein SF, Way LW. Surgical treatment of chronic pancreatitis. Am J Surg 1984;148:754-9.  Back to cited text no. 8    
9.Tripodi AM, Sherwin CF. Experimental transplantation of the pancreas into the stomach. Arch Surg 1934;28:345-56.  Back to cited text no. 9    
10.Thomas PG, Augustine P, Ramesh H, Rangabashyam N. Observation and surgical management of tropical pancreatitis in Kerala and Southern India. World J Surg 1990;14:32-42.  Back to cited text no. 10    
11.Madsen P, Hansen E. Coeliac plexus block versus pancreaticogastrostomy for pain in chronic pancreatitis. A controlled randomized trial. Scand J Gastroenterol 1985;20:1217-20.  Back to cited text no. 11    
12.Efron DT, Lillemoe KD, Cameron JL, Yeo CJ. Central pancreatectomy with pancreaticogastrostomy for benign pancreatic pathology. J Gastrointest Surg 2004;8:532-8.  Back to cited text no. 12    
13.Kovacs I, Toth P, Arkossy P, Sapy P. Changes in the gastric acidity after ductal decompression in patients with chronic pancreatitis. Eur J Surg 1999;165:786-90.  Back to cited text no. 13    
14.Ebbehψj N, Klaaborg KE, Kronborg O, Madsen P. Pancreaticogastrostomy for chronic pancreatitis. Am J Surg 1989;157:315-7.  Back to cited text no. 14    
15.Bassi C, Falconi M, Tihany T, Salvia R, Valerio A, Caldiron E, et al . Resection in chronic pancreatitis: Anastomosis with the jejunum or with the stomach? Ann Ital Chir 2000;71:51-5.  Back to cited text no. 15    
16.Amano H, Takada T, Ammori BJ, Yasuda H, Yoshida M, Uchida T, et al . Pancreatic duct patency after pancreaticogastrostomy: Long term follow-up study. Hepatogastroenterology 1998;45:2382-7.  Back to cited text no. 16    
17.Bradley EL 3 rd . Long term results of pancreaticojejunostomy in patients with chronic pancreatitis. Am J Surg 1987;153:207-13.  Back to cited text no. 17    
18.Sato T, Miyashita E, Yamauchi H, Matsuno S. The role of surgical treatment of chronic pancreatitis. Ann Surg 1986;203:266-71.  Back to cited text no. 18    
19.Wilson T, Hollands M, Little J. Pancreaticojejunostomy for chronic pancreatitis. Aust N Z J Surg 1992;62:111-5.  Back to cited text no. 19    
20.Holmberg JT, Isaksson G, Ihse I. Long term results of pancreaticojejunostomy in chronic pancreatitis. Surg Gynaecol Obstet 1985;160:339-46.  Back to cited text no. 20    
21.Sielezneff I, Malouf A, Salle E, Brunet C, Thirion X, Sastre B. Long term results of lateral pancreaticojejunostomy for chronic alcoholic pancreatitis. Eur J Surg 2000;166:58-64.  Back to cited text no. 21    

Correspondence Address:
Prosanta K Bhattacharjee
Flat No. 10-C, 9, Mandeville Gardens, Kolkata - 700 019
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/1319-3767.32180

Rights and Permissions


  [Figure - 1], [Figure - 2], [Figure - 3]

  [Table - 1], [Table - 2]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Email Alert *
    Add to My List *
* Registration required (free)  

    Materials and Me...
    Article Figures
    Article Tables

 Article Access Statistics
    PDF Downloaded531    
    Comments [Add]    

Recommend this journal