The aim of this study is to compare the closure rate of pancreatic fistula (PF) after pancreaticoduodenectomy (PD) under various types of nutrition.

Postoperative pancreatic fistula (POPF) is the most detrimental complication of pancreatic surgery due to the potential life-threatening consequences of fluid and electrolyte imbalance, nutritional depletion, sepsis, and local complications, such as abdominal abscess and hemorrhage.1 Incidence rates vary considerably from 0 to 24% due to the differences in definitions with an overall rate of 12.9% in a recent series.1,2 Several risk factors have been identified, mostly a soft parenchyma and a small main pancreatic duct diameter; however, numerous attempts to reduce fistula rates with either pharmacological prophylaxis or application of some special surgical techniques failed to improve postoperative outcomes. The primary therapy for POPF today includes adequate drainage of pancreatic exocrine secretions and conservative treatment, consisting of nutritional support and correction of electrolyte disturbances and fistula-related complications. 3-5 These recommendations are based on previous observations suggesting that up to 70% of cases eventually resolve spontaneously.3,5,6 However, the patient's discomfort, the need for follow-up visits and the substantial costs of prolonged therapy initiated various attempts to accelerate closure rates. Several measures have been proposed, including fibrin glues, endoscopic interventions, or the use of somatostatin analogues to inhibit pancreatic exocrine secretion. Nevertheless, these methods lack any convincing evidence that the proposed regimens may be clinically effective.4,7,8 Nutritional support is the key element of conservative therapy in patients with POPF, as most of them are in a catabolic state and attempts to accelerate fistula closure usually involve prolonged fasting. However, the decision between total parenteral nutrition (TPN) and enteral nutrition is essentially arbitrary because the effects of both diets on closure rates of postoperative pancreatic fistula have not been compared in a randomized clinical trial.3,9 Experiments with healthy individuals have demonstrated that intravenous feeding does not stimulate pancreatic secretion, and thus is a reasonable solution when prolonged nutritional support is needed without increasing the exocrine pancreatic function.10 However, previous research has suggested that long-term TPN leads to negative functional and morphological changes, not only within the gastrointestinal mucosa but also atrophy and dysfunction of the exocrine pancreas.11 Therefore, enteral feeding beyond the ligament of Treitz is commonly preferred over the intravenous route due to lower costs and the potential advantage of avoiding infectious and metabolic complications related to the parenteral route. This hypothesis was substantiated in several clinical trials demonstrating that enteral nutrition via a nasojejunal tube can be safely used in patients with various disorders, including acute pancreatitis and postoperative pancreatic fistula. 12-15Some of these reports suggested that the enteral route offers major advantages over TPN in terms of faster recovery and lower rates of disease- and nutrition-related complications.16 However, data concerning chronic pancreatic conditions are limited.17 It is well established that the duodenum is the major site of pancreatic secretion stimulation. Cholecystokinin (CCK) and secretin released in the duodenum and enteropancreatic reflexes mediated by vago-vagal cholinergic pathways are responsible for the majority of pancreatic exocrine secretion.18Without raising CCK levels, the enteropancreatic reflex can be activated with a corresponding increase in pancreatic enzyme secretion, by distention or administration of hyperosmolar solutions in the duodenum. Pancreatic polypeptide (PP) secretion is also under cholinergic control and thus may be a modulator of pancreatic secretion stimulated by the vagal cholinergic pathway.19 It has also been shown that intraileal or colonic perfusion of nutrients decreases pancreatic exocrine secretion, possibly mediated through the ileal-brake gut peptides, namely peptide YY (PYY) and glucagon-like peptide-1 (GLP-1). 20-22 Enteral feeding via a nasojejunal feeding tube necessitates endoscopic placement which could be dangerous in immediately operated patients with gastrojejunostomy and might cause tube-related discomfort to patients. Anatomically, pancreaticoduodenectomy will include removal of duodenum and proximal 10 ~ 15 cm jejunum (Figure 1). In addition, another 30~40 cm-long jejunum will be brought up for pancreatic and biliary anastomosis (figure 2). Therefore, the last enteral anastomosis (gastrojejunostomy or duodenojejunostomy) will be made at site of 40~50 cm distal to Treitz ligament which is far more distal than site of nasojejunal tube used in patients with acute pancreatitis. Therefore, we hypothesize that oral feeding in patients operated with PD will not stimulate but inhibit pancreatic secretion and hasten closure of pancreatic fistula. To test this hypothesis, we propose a prospective randomized trial to test the effect of various nutrition methods on healing of PF after PD. Patients will be randomized into 2 groups (A and B). Eligible patients will be randomized in a 1:1 ratio to receive oral feeding nutrition, or TPN as a standard therapy of POPF. The primary end points will be closure rate of PF. Secondary end points will include length and cost of hospital stay after operation.

pancreatic exocrine secretion. Nevertheless, these methods lack any convincing evidence that the proposed regimens may be clinically effective.

Nutritional support is the key element of conservative therapy in patients with POPF, as most of them are in a catabolic state and attempts to accelerate fistula closure usually involve prolonged fasting. However, the decision between total parenteral nutrition (TPN) and enteral nutrition is essentially arbitrary because the effects of both diets on closure rates of postoperative pancreatic fistula have not been compared in a randomized clinical trial. Experiments with healthy individuals have demonstrated that intravenous feeding does not stimulate pancreatic secretion, and thus is a reasonable solution when prolonged nutritional support is needed without increasing the exocrine pancreatic function. However, previous research has suggested that long-term TPN leads to negative functional and morphological changes, not only within the gastrointestinal mucosa but also atrophy and dysfunction of the exocrine pancreas. Therefore, enteral feeding beyond the ligament of Treitz is commonly preferred over the intravenous route due to lower costs and the potential advantage of avoiding infectious and metabolic complications related to the parenteral route. This hypothesis was substantiated in several clinical trials demonstrating that enteral nutrition via a nasojejunal tube can be safely used in patients with various disorders, including acute pancreatitis and postoperative pancreatic fistula. Some of these reports suggested that the enteral route offers major advantages over TPN in terms of faster recovery and lower rates of disease- and nutrition-related complications. However, data concerning chronic pancreatic conditions are limited.

It is well established that the duodenum is the major site of pancreatic secretion stimulation. Cholecystokinin (CCK) and secretin released in the duodenum and enteropancreatic reflexes mediated by vago-vagal cholinergic pathways are responsible for the majority of pancreatic exocrine secretion. Without raising CCK levels, the enteropancreatic reflex can be activated with a corresponding increase in pancreatic enzyme secretion, by distention or administration of hyperosmolar solutions in the duodenum. Pancreatic polypeptide (PP) secretion is also under cholinergic control and thus may be a modulator of pancreatic secretion stimulated by the vagal cholinergic pathway. It has also been shown that intraileal or colonic perfusion of nutrients decreases pancreatic exocrine secretion, possibly mediated through the ileal-brake gut peptides, namely peptide YY (PYY) and glucagon-like peptide-1 (GLP-1).

Enteral feeding via a nasojejunal feeding tube necessitates endoscopic placement which could be dangerous in immediately operated patients with gastrojejunostomy and might cause tube-related discomfort to patients. Anatomically, pancreaticoduodenectomy will include removal of duodenum and proximal 10 ~ 15 cm jejunum (Figure 1). In addition, another 30~40 cm-long jejunum will be brought up for pancreatic and biliary anastomosis (figure 2). Therefore, the last enteral anastomosis (gastrojejunostomy or duodenojejunostomy) will be made at site of 40~50 cm distal to Treitz ligament which is far more distal than site of nasojejunal tube used in patients with acute pancreatitis. Therefore, we hypothesize that oral feeding in patients operated with PD will not stimulate but inhibit pancreatic secretion and hasten closure of pancreatic fistula. To test this hypothesis, we propose a prospective randomized trial to test the effect of various nutrition methods on healing of PF after PD. Patients will be randomized into 2 groups (A and B). Eligible patients will be randomized in a 1:1 ratio to receive oral feeding nutrition, or TPN as a standard therapy of POPF. The primary end points will be closure rate of PF. Secondary end points will include length and cost of hospital stay after operation.

Inclusion Criteria:

• aged more than 20 years old and develop pancreatic fistula after pancreatectomy

Exclusion Criteria:

• major co-morbidities
• severe pancreatic fistula needing intervention