Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Gastroduodenal surgery is often performed for severe peptic ulcer disease (PUD), benign or malignant neoplasms, and obesity. Radiologic evaluation of the postoperative stomach and duodenum requires an understanding of the surgery performed, expected postoperative appearances, and potential complications. During the early postoperative period, the radiologist is often asked to evaluate the surgical anatomy and to assess for complications such as leaks. During the late postoperative period, the radiologist may be asked to evaluate symptoms such as nausea, vomiting, abdominal pain, and weight loss or gain and to assess for delayed complications. This chapter provides an overview of the indications for gastroduodenal surgery, approaches for radiologic evaluation, and expected postoperative appearances and complications on imaging studies.
The imaging approach after surgery on the stomach and duodenum depends on the surgical procedure, postoperative time course, and clinical status of the patient. Radiologic evaluation usually entails fluoroscopic examinations or computed tomography (CT), and both imaging studies may be needed for accurate diagnosis of postoperative complications.
A scout radiograph should always be obtained before an upper gastrointestinal (GI) study is performed on postoperative patients to assess for surgical material or other radiopaque densities that could be mistaken for extraluminal collections after contrast medium is administered. The pattern of sutures or staples in the surgical bed may also help determine the nature of the surgery when the clinical history is inadequate. Expected (e.g., gastric bands) or unexpected (e.g., sponges) surgical foreign bodies may also be assessed.
When gastroduodenal surgery is performed, each procedure has many variations, so a “cookbook” approach for the upper GI study is generally inadequate for evaluating these patients. Instead, the radiologist needs to modify examination technique and patient positioning for each patient based on the surgical anatomy. It is particularly important for the initial contrast bolus to be carefully observed and promptly imaged at fluoroscopy as it traverses an area of resection, bypass, or banding before opacification of overlapping structures obscures strictures, leaks, or other findings at the surgical site.
Early postoperative upper GI studies should be performed with water-soluble contrast agents to assess for leaks, as these agents have no known deleterious effects if they extravasate into the mediastinum or peritoneal cavity. If no leak is identified, barium should then be administered to detect small or subtle leaks that could be missed with water-soluble contrast agents. In contrast, barium should be administered as the initial contrast agent for upper GI studies performed during the late postoperative period (assuming no leak is suspected). Recumbent positioning may help to distend the postoperative structures being assessed and is often the only way to evaluate sick or debilitated patients.
When upper GI studies are performed after gastroduodenal surgery, single-contrast technique is preferable to double-contrast technique for assessing the surgical anatomy, especially anastomoses, and to evaluate for leaks, fistulas, strictures, or staple line dehiscence. Double-contrast studies also are difficult to perform on postoperative patients because of their limited mobility.
CT provides detail about the entire abdomen and pelvis, including the bowel in its entirety, and enables detection of intra-abdominal collections. CT is ideally performed utilizing positive oral and intravenous (IV) contrast material. Positive oral contrast facilitates depiction of the surgical anatomy. A water-soluble oral contrast agent should be utilized during the early postoperative period or any time after surgery if perforation is suspected. An oral contrast agent is often given to the patient at intervals prior to CT and should also be administered immediately before image acquisition.
Nuclear medicine studies are performed to evaluate gastric emptying, motility problems, or postgastrectomy abnormalities such as GI stasis and dumping syndrome.
Endoscopy is performed to assess mucosal abnormalities, particularly at surgical anastomoses, and endoscopic biopsy specimens enable detection of histologic findings associated with bile reflux gastritis, marginal ulcers, or recurrent tumor.
Surgery for PUD has dramatically declined since the 1990s because of greater use of effective acid-suppressive agents and widespread treatment of Helicobacter pylori . Surgery is usually reserved for severe or emergent complications.
Indications for surgical management of PUD include bleeding, perforation, obstruction, and intractable ulcers. , GI bleeding is the most frequent and severe complication, with a mortality rate approaching 10%. , The use of various endoscopic and interventional radiology procedures has decreased the need for surgery in patients with bleeding ulcers, but surgical intervention may still be required for treatment of shock or persistent or recurrent GI bleeding despite optimal nonsurgical management. Perforated ulcers represent a surgical emergency, with morbidity and mortality rates as high as 50% and 30%, respectively. , , Gastric outlet obstruction or duodenal obstruction from PUD-related scarring and nonhealing ulcers on protracted medical therapy may also necessitate surgery. ,
A minimalistic surgical approach to PUD includes simple oversewing of ulcers, omental patch repair, ulcerectomy, or ligation of nearby vessels. , More definitive approaches entail techniques for decreasing acid secretion from gastric parietal cells, often combined with resection of the gastric mucosa at highest risk for recurrent ulcers. Such procedures include truncal vagotomy with pyloroplasty, vagotomy with distal gastrectomy, and highly selective vagotomy (HSV).
Vagotomy decreases total acid secretion and parietal cell response to gastrin and other stimulants. Truncal vagotomy requires complete transection of the anterior and posterior vagus nerve trunk with denervation of the stomach, liver, gallbladder, pancreas, small bowel, and proximal colon. Loss of the antropyloric mechanism results in gastric stasis, so a concomitant pyloroplasty, gastroduodenostomy, or gastrojejunostomy may be performed to improve gastric drainage. Truncal vagotomy causes dumping or diarrhea in 10% of patients.
HSV is a relatively safe procedure that decreases gastric acid secretion by 75%. Unlike patients who undergo nonselective vagotomy, distal branches of the vagal nerve supply to the antrum and pylorus are preserved, so the antropyloric mechanism remains intact. Because gastric emptying is unaltered, a surgical drainage procedure, resection, or anastomosis is no longer required. Nevertheless, elective HSV has largely been replaced by long-term medical treatment of ulcers with proton pump inhibitors. ,
Pyloroplasty is a gastric drainage procedure that entails reconstruction and enlargement of the pyloric channel to improve gastric emptying after the normal antropyloric mechanism is interrupted by truncal vagotomy. Pyloroplasty is relatively easy to perform and less invasive than antrectomy, preserving upper GI continuity.
Antrectomy removes the source of gastrin secretion and prevents gastric stasis. When combined with truncal vagotomy, antrectomy is the gold standard for reducing acid secretion and is associated with lower rates of ulcer recurrence (<1%) than vagotomy with pyloroplasty (4% to 27%) or HSV alone (4% to 11%). , The lower recurrence rate of ulcers after antrectomy and truncal vagotomy must be weighed against a higher rate of complications, including diarrhea, dumping, bile reflux gastritis, weight loss, and metabolic disorders. , Gastric resection also entails a higher morbidity and mortality than HSV alone or vagotomy with pyloroplasty. ,
An antrectomy necessitates creation of a surgical anastomosis at the gastric outlet. Either a gastroduodenostomy (the more physiologic anastomosis) or gastrojejunostomy may be performed. The latter anastomosis requires closure of the duodenal stump.
Partial gastrectomy entails resection of the antrum, distal two- thirds or four-fifths of the stomach, or subtotal gastrectomy, depending on the nature and location of disease. Restoration of GI continuity requires creation of an anastomosis at the gastric outlet. The procedure is named based on the type of anastomosis between the gastric remnant and small bowel, regardless of the extent of resection. Procedures associated with a distal gastric resection include Billroth I, Billroth II, and Roux-en-Y gastrojejunostomy.
When an antrectomy is performed for PUD, GI continuity is reestablished with Billroth I gastroduodenostomy or Billroth II gastrojejunostomy. When an antrectomy is performed for ulcer disease, there is usually a large gastric remnant consisting of 60% to 70% of the stomach. Roux-en-Y gastrojejunostomy should be avoided because a large gastric remnant predisposes to gastric stasis and marginal ulcers. In contrast, a distal gastrectomy for neoplasms leaves a smaller gastric remnant, so Roux-en-Y gastrojejunostomy may be preferable to Billroth II in this setting for decreasing reflux of biliopancreatic secretions into the gastric remnant and esophagus.
The Billroth I procedure consists of an antrectomy with a gastroduodenal anastomosis. This procedure entails anastomosis of the distal end of the gastric remnant to the proximal duodenum to restore upper GI continuity ( Fig. 23.1 ). The gastroduodenal anastomosis should approximate the size of the pylorus to delay gastric emptying and reduce the frequency of postgastrectomy dumping. Because of anastomotic requirements, gastroduodenostomy can only be performed after a distal antrectomy, whereas gastrojejunostomy is usually performed for more substantial gastric resections.
The Billroth II procedure consists of a distal gastrectomy with loop-type gastrojejunostomy ( Fig. 23.2 ). A loop of jejunum 12 to 15 cm distal to the ligament of Treitz is anastomosed to the greater curvature of the gastric remnant, creating an afferent loop (the duodenal stump and proximal jejunum) carrying biliopancreatic secretions and an efferent loop carrying gastric contents into the small bowel.
Roux-en-Y gastrojejunostomy consists of jejunal transection with the proximal end of the jejunum distal to the transection anastomosed to the gastric remnant and the distal end of the jejunum proximal to the transection anastomosed to the small bowel at least 40 to 60 cm distal to the gastrojejunal anastomosis ( Fig. 23.3 ). This procedure diverts biliopancreatic and duodenal secretions away from the stomach, helping to prevent bile reflux gastritis. Nevertheless, jejunal transection may decrease intestinal motility, impairing emptying of the Roux limb (the limb extending from the duodenal stump to the jejunojejunal anastomosis). Ideally, a small gastric remnant is created to help avoid gastric stasis and prevent the development of marginal ulcers at the gastrojejunal anastomosis.
The surgical anatomy is readily identified on upper GI studies after distal gastrectomy procedures. In gastroduodenostomy, there is foreshortening of the stomach and absence of a normal pylorus with an end-to-end gastroduodenal anastomosis (see Fig. 23.1 ). This procedure is difficult to recognize on CT if the majority of the stomach remains intact and no radiopaque clips or staples are identified. In contrast, a gastrojejunostomy can be recognized on CT when it shows a jejunal limb contiguous with the gastric remnant (see Fig. 23.2D ) and radiopaque staples are often seen at the proximal duodenal stump.
In gastrojejunostomy, an upper GI study typically shows surgical absence of the antrum with a widely patent gastrojejunal anastomosis (see Figs. 23.2B and C ). The cut end of the stomach is often oversewn or inverted to restrict stomal size, producing plication defects at the anastomosis that can be mistaken for leak, ulcers, or recurrent tumor on fluoroscopic studies. , However, a plication defect should fill and empty at fluoroscopy, whereas contrast material usually remains in a leak or ulcer after passage of the bolus. With marginal ulcers or recurrent tumor, adjacent gastric folds are often thickened and lobulated. When the findings are equivocal, endoscopy may be required for a definitive diagnosis.
Gastric resection sometimes results in iron-deficiency anemia because of inadequate oral intake, malabsorption, or chronic GI bleeding. Decreased production of intrinsic factor in the gastric remnant can also lead to vitamin B 12 deficiency, and impaired calcium absorption in the duodenum can lead to metabolic disease. Afferent loop syndrome and malabsorption of fat related to insufficient mixing of food with digestive enzymes may also result in decreased absorption of fat-soluble vitamin D.
The frequency of anastomotic leaks after gastric resection is about 5%. Leaks typically occur at the gastroduodenal anastomosis after Billroth I, at the gastrojejunal anastomosis or oversewn duodenal stump after Billroth II, and at the gastrojejunal or jejunojejunal anastomosis or duodenal stump after Roux-en-Y gastrojejunostomy.
Postoperative leaks are characterized on upper GI studies by focal extravasation of contrast material into contained extraluminal tracks or collections ( Fig. 23.4 ) or by free extravasation of contrast material into the peritoneal cavity. Surgical drains should be carefully evaluated, as some leaks may be recognized only by opacification of the drain with contrast material. Dehiscence of the duodenal stump is a particularly serious complication, resulting in leakage of biliopancreatic sections into the peritoneal cavity. Unfortunately, the duodenal stump may not be adequately visualized on upper GI studies because of incomplete retrograde filling of the afferent limb and duodenum after gastrojejunostomy. Nevertheless, duodenal stump leaks can still be diagnosed on CT showing complex fluid- and/or gas-containing collections abutting the stump ( Fig. 23.5 ).
Gastric stasis after partial gastrectomy occurs in the absence of stenosis or obstruction in 25% of patients. This complication causes postprandial bloating, vomiting, abdominal pain, and weight loss. Ineffective gastric emptying, impaired intestinal motility, and alkaline reflux gastritis all contribute to gastric stasis. An upper GI study usually shows a widely patent anastomosis without evidence of obstruction. Nuclear medicine studies are sometimes helpful for assessing gastric emptying. Severe gastric stasis may necessitate surgical revision (with repositioning of the anastomosis) or even a total gastrectomy.
Roux stasis syndrome may occur after Roux-en-Y gastrojejunostomy. There is usually an atonic gastric remnant with delayed emptying of the remnant and jejunal Roux limb. , This condition is thought to represent a motility disorder. Patients present with nausea, vomiting, abdominal pain, and weight loss. An upper GI study may show a dilated gastric remnant and Roux limb with delayed emptying but no evidence of mechanical obstruction. The jejunojejunal anastomosis should be carefully evaluated to rule out a more distal anastomotic stricture as the cause of these findings. Nuclear medicine scans may also show delayed emptying of the gastric remnant and Roux limb. Similarly, GI motility testing may show abnormal motility in the Roux limb with reverse propulsive activity toward rather than away from the stomach. Promotility agents are beneficial for some patients.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here