The first successful pancreaticoduodenectomy was performed in 1912. Whipple popularized the procedure for periampullary tumors, reporting three cases in 1935. The original operation was performed in two stages. The morbidity and mortality rates after pancreaticoduodenectomy were high for several decades, and long-term survival was extremely poor. Since that time, numerous modifications and technical refinements have been made, and perioperative mortality rates have dropped from approximately 20% to less than 4% [37,38]. The "standard" Whipple procedure now consists of resection of the head and uncinate process of the pancreas, the common bile duct and gallbladder, and the duodenum, with formation of three anastomoses: a choledochojejunostomy, a gastrojejunostomy, and a pancreatico-jejunostomy (Fig. 3.40).
Variations in the type and placement of the anastomoses are common. The jejunal loop may be placed in an antecolic or retrocolic position. Stents are sometimes left in the pancreatic duct and the chole-dochojejunal anastomosis temporarily after surgery. In the pylorus-preserving Whipple, which has become an alternative to the standard procedure, the pylorus and first portion of the duodenum are preserved (Fig. 3.41). This modification results in better postoperative weight gain due to the increased gastric reservoir, decreased bile reflux, and decreased jejunal ulceration (since gastrointestinal physiology is more normal). The operative time is also shortened. There may
however be a temporary delay in gastric emptying. Indications for the Whipple procedure include neoplasms of the periampullary region, symptomatic chronic pancreatitis, and trauma.
The evaluation of the postoperative Whipple patient is challenging because of the complexity of the surgery and the multiple anastomoses to be investigated. In the early postoperative period UGI series and CT are the most useful tools available, with CT imaging being preferable in the later postoperative period (Figs. 3.42 and 3.43). The CT features of the normal postoperative appearance and complications have been described by multiple authors [39-44]. Attempts should be made to identify the three anastomoses. The pancreaticojejunostomy and chole-dochojejunostomy may be difficult to identify because they are often not opacified with oral contrast medium, although if stents are placed across these anastomoses, they can generally readily be identified. Pneumobilia is seen in about 80% of patients; thus its absence does not imply obstruction of the choledochojejunal anastomosis . Transient fluid collections in the surgical bed, in Morrison's pouch, in the right paracolic gutter, and in the region of the anastomoses are common findings but may be difficult to differentiate from leaks and abscesses . When fluid collections are seen on CT scans, an UGI series with water-soluble contrast may be necessary to exclude a leak, and ultrasound-or CT-guided aspiration may be necessary to differentiate abscess from noninfected fluid. Reactive lymphadenopathy may be a normal inflammatory response to recent surgery and may be confused with metasta-tic disease. Serial scans are helpful in this regard, since reactive nodes
Was this article helpful?