Gastric remnant obstruction due to adenocarcinoma after remote roux-en-y gastric bypass

Case contributed by Dr Michael P Hartung


History of roux-en-y gastric bypass 40 years ago. 5 days of worsening abdominal pain and distension.

Patient Data

Age: 70 years
Gender: Female

Patent antecolic Roux-en-Y gastric bypass with oral contrast within multiple loops of small bowel in the mid abdomen. 

Moderate to severe distention of the gastric remnant containing mostly fluid. Area of thickening or possible twisting of the gastric antrum (best appreciated on the coronal reformats). 

Remainder of the pancreaticobiliary limb appears normal in caliber.

Dilated gastric remnant thought to be secondary to gastric volvulus or adhesions, 40 years status post Roux-en-Y gastric bypass.


  1. Exploratory laparotomy with lysis of adhesions greater than 1 hour.
  2. Gastric decompression.
  3. Gastroscopy with biopsy of antral mucosa.
  4. Placement of Stamm gastrostomy utilizing a 26-French Malecot tube.


....Immediately upon entering the peritoneal cavity, the dilated gastric remnant bulged into the operative field. There was densely adherent overlying omentum and as I was trying to manually free the stomach in order to attempt to understand the nature of the obstruction, we encountered some blood loss from some omental vessels laterally. The decision was made to decompress the stomach. A purse-string suture of 3-0 silk was created and a stab incision was made on the anterior aspect of the stomach. A Poole sucker was inserted and there was immediate return of at least 1000 cc of non-bilious, grey-tinged, clear fluid. The Poole sucker was removed and the purse-string closed. Bleeding was controlled with suture ligation as well as clamping and tying any actively bleeding vessels. Once the bleeding was under control, the area was irrigated, suctioned of all irrigant, the process of lysing adhesions was carried out in a more formal fashion, first dividing the adhesions between the omentum and the anterior abdominal wall laterally on either side of the incision as well as inferiorly. Once the entire omentum was mobilized and the transverse colon was brought inferior, the stomach was well visualized. At this point, there did not appear to be any residual volvulus of the gastric remnant. Whether this was resolved with decompression is difficult to determine, as the stomach itself was not completely visualized until after greater than 1 hour of adhesiolysis. Nevertheless, this was palpated to scar tissue laterally on the left until adhesions in the area of the original division of the gastric cardia from the fundus were encountered. The Roux limb appeared viable and nondistended up to the gastric remnant. This was left alone. Palpation distally was significant for thickening and induration in the pyloric region. I was concerned about the possibility of a mass in this area, as there was some serosal adhesions in this area from omentum that ultimately extended towards the falciform ligament. A remnant of the falciform ligament and some of this omentum had to be clamped, divided and excised and removed. It was sent for permanent histopathologic evaluation. The pylorus felt indurated and thickened. It was difficult to tell if this was scar tissue from longstanding adhesions that had ultimately caused an obstruction or a thickening secondary to a previous torsion that had now let up once the stomach was decompressed. At this point, the gastrostomy was reopened by dividing the purse-string suture and a 12-French red rubber catheter was inserted and advanced through the pyloric channel. It appeared that the gastric remnant was in alignment and that it was now patent, at least to 12-French, as there was return of bile. Recall that the fluid removed from the stomach at the time of decompression was nonbilious. At this point a sterile cystoscope was brought onto the field and inserted through the gastrostomy, secured with purse-string suture once again. The strands were left long and held with a clamp, such that they could be used to purse-string around the cystoscope. The gastric remnant was insufflated with air and the scope advanced to what appeared to be the level of the antrum. We clamped the duodenum with an atraumatic clamp distally to prevent air escape into the small bowel, and we insufflated the stomach. The antral mucosa appeared normal. It was photographed. There was some bruising, most likely from manipulation and advancement of the red rubber catheter. Using a cold forceps, a biopsy was obtained of the antral mucosa. This was sent for permanent histopathologic evaluation. We could not adequately visualize the pyloric channel or beyond the pylorus. The scope was removed. The abdomen was irrigated, suctioned of all irrigant, and evaluated for blood loss. There were some areas of clotted blood in the suprahepatic space bilaterally. This was irrigated and suctioned until clear. The Stamm gastrostomy was created by lining up our gastrostomy opening on the anterior wall of the stomach with a position on the anterior abdominal wall. An overlying skin incision was made. A hemostat was used to create an opening and a 26-French Malecot tube was placed through the abdominal wall and into the gastrostomy and the purse-string was secured. Four box sutures were placed to seal the anterior gastric wall to the overlying abdominal wall with airtight seal between the peritoneum and the serosa of the stomach. These were secured and the Malecot was brought out such that there was no tension on the gastric wall, but it was snug to the gastrostomy. It was secured in place with nylon suture. The abdomen was again irrigated, suctioned of all irrigant. A single sheet of Seprafilm was placed after the omentum was brought down over the midline and the abdomen was closed with interrupted 0 PDS sutures in a Smead-Jones closure. Skin edges were reapproximated with skin clips. Antibiotic ointment and a sterile gauze were applied to the incision and a drain sponge applied at the gastrostomy site. At the conclusion of the case, John Jarocki placed a TAP block. The patient tolerated the procedure well and there were no immediate postoperative complications. 



Part "A" - Gastric antrum, biopsy - Antral mucosa without acute inflammation, special stain
for Helicobacter pylori is negative.
Part "B" - Remnants of falsiform ligament, excision - Fibromembranous and adipose tissue
with acute and chronic inflammation.

Ultrasound guided biopsy of pylorus was performed due to persistent obstruction:



RESULTS: Malignant Cells Present Derived From Adenocarcinoma.

Immunoperoxidase Stain(s): Cytokeratin Cocktail And CK7 Positive; CK20, S100

REMARKS: Morphology And Immunophenotype Would Favor An Upper GI Tract Primary.

Case Discussion

This case provides a grade example of an important consideration when evaluating a gastric bypass patient with acute abdominal pain: evaluate the gastric remnant and pancreaticobiliary limb carefully. 

Because the gastric remnant and pancreaticobiliary (afferent) limb is excluded from the efferent gastrointestinal tract (flow of ingested material), obstruction will not result in typical vomiting and relief of pressure. Instead, these patients will have worsening, severe abdominal pain as in this case, and can eventually proceeded to ischemic necrosis and perforation.

A gastric bypass patient should never such dilation of the blind-ending stomach as seen in this case, without raising the suspicion for obstruction of the afferent loop. In this case, there is thickening and angulation of the gastric antrum which was thought to represent volvulus of the gastric remnant (see coronal reformats). A mass was not suspected on the initial interpretation. 

At surgery, greater than one hour of adhesiolysis was performed, suggesting adhesions of the cause rather than volvulus. The operative report is quite long but provides a lot of detail and insight into the case. Surgical biopsy was negative for cancer. 

The patient had a venting g-tube placed at surgery and failed clamping, raising the concern for a mass causing the obstruction. Ultrasound guided biopsy was performed after filling the obstructed stomach with fluid, allowing a clear acoustic window for biopsy of the site of obstruction at the pylorus. Pathology results showed adenocarcinoma as responsible for the obstruction!

It is also important to evaluate for dilation of the duodenum and promixal jejunum to the jejunojejunostomy, as small bowel stricture or volvulus distally could also occur (afferent loop syndrome). These ares are normal in this case. 

How to use cases

You can use Radiopaedia cases in a variety of ways to help you learn and teach.

Creating your own cases is easy.

Updating… Please wait.

 Unable to process the form. Check for errors and try again.

 Thank you for updating your details.