Pancreaticoduodenectomy (PD) includes en bloc resection of duodenum, head of pancreas, proximal jejunum, distal common bile duct (CBD), gallbladder, and distal stomach, which was followed by complicated reconstructions. Therefore, PD causes change of physiological functions, containing insufficiency of exocrine and endocrine pancreatic function, malabsorption of nutrients, impairment of gut peristalsis, and hepatic steatosis . On the other hand, DM may also resolve after PD in cases with pancreatic ductal adenocarcinoma. From the study of bariatric operation, resolution of DM may occur even before body weight loss. Postulated mechanism of resolution of DM after bariatric operation included foregut and hind gut theory. Actually, both hindgut and foregut effects on sugar control are achieved to some extent after pancreaticoduodenectomy. First, most of the duodenum is routinely removed in PD and there will be no food passage through duodenum (foregut theory) after PD. Second, proximal 10 cm of jejunum will be removed in PD and another 30~ 40 cm- long jejunum will be brought up for pancreatic and biliary anastomosis, which will make the last enteral anastomosis (gastrojejunostomy or duodenojejunostomy) be created at site about 50-60 cm distal to Treitz ligament. Therefore, food will directly pass into distal jejunum (hindgut theory). In our preliminary study, 35% patients have resolution of diabetes after PD, especially for ones with new-onset DM. Further, a prospective randomized clinical trial will be conducted to address if modified distal gastro-/duodeno-jejunostomy results in a higher proportion of diabetes remission compared with standard PD.

All new-onset DM patients undergoing PD will be considered for inclusion in this prospective randomized study. Inclusion criteria will be age greater than 20 years and planned PD for a lesion of either the pancreatic head or the periampullary region. The patients received standard PD are defined as control group; the cases receiving more distal gastro-/duodeno-jejunostomy ( 60 cm distal to gastro-/duodeno-jejunostomy on the control group) are defined as study group . The trial will be conducted only after obtaining an approval for the study design from the National Taiwan University institutional ethics review board. The type of surgery (pylorus-preserving or standard PD) and the type of management of the pancreatic stump (pancreaticojejunostomy or pancreaticogastrostomy) will be left to the surgeons' discretion. This trial will be done for two years to investigate if distal enteral bypass results in more remission of diabetes in new-onset DM cases after PD .After PD, FBG and HbA1c were routinely checked every 3 months for two years. The definition of remission of DM was that a fasting glucose level < 110 mg/dL and HbA1c < 6.0% without the use of oral hypoglycemic agents or insulin was defined as complete remission. A fasting glucose value < 126 mg/dL and HbA1c <6.5% was defined partial remission.32 In addition, the body weight will be checked regularly. The demographic, pathological, and clinical data will be recorded. The preoperative and postoperative antidiabetic medication and doses were also included.

In plan (to study if more distal bypass of gastrojejunostomy may increase remission of new-onset DM cases), the major concern of potential risk for patients participating the clinical trial is malnutrition due to enteral bypass. Other surgical procedures are the same as standard PD. For the monitoring of malnutrition, patients visited outpatient clinic every month for the first six months, and every three months afterward. The complete blood count and value of serum albumin will be checked very three months. The body weight loss > 10% during three months is considered malnutrition. The peripheral parenteral nutrition will be administered to manage the malnutrition, and total parenteral nutrition should be used if malnutrition persisted. However, malnutrition is also a common phenomenon for recurrent cancer cases. The management of malnutrition for cancer patients will be managed according to clinical guidelines at National Taiwan University Hospital.

Inclusion Criteria:

• DM
• Patients with impaired fasting glucose before operation

Exclusion Criteria:

• Type 1 diabetes
• History of pancreatitis
• Hepatic dysfunction (Child-Pugh > 2)
• Renal dysfunction (serum creatinine concentration > 3 mg/L, hemodialysis, or both)
• Pregnancy
• Malnutrition (pre-operative BMI <20)