Prophylactic Total Gastrectomy for CDH1 Gene Mutation

The Hereditary Diffuse Gastric Cancer (HDGC) syndrome is due to a mutation in the CDH1 gene that predisposes patients to a high lifetime risk of developing gastric cancer. As such, a total gastrectomy is typically recommended for patients with this syndrome. In this case, the patient presented with an incidentally discovered CDH1 mutation on genetic testing obtained after she was diagnosed with early-onset rectal cancer. She underwent a prophylactic total gastrectomy with a retrocolic Roux-en-Y esophagojejunostomy. Her postoperative course was unremarkable, and she was discharged on postoperative day 3. Her pathology demonstrated several foci of signet ring cell carcinoma limited to the mucosa. This video demonstrates an experienced surgeon’s technique for performing a prophylactic total gastrectomy with a Roux-en-Y esophagojejunostomy reconstruction.

Hereditary diffuse gastric cancer (HDGC) syndrome is an autosomal dominant genetic condition due to a mutation in the CDH1 gene. It is a rare condition, responsible for only 1–3% of all gastric cancers. However, patients with this condition have a 56–83% lifetime risk of developing gastric cancer and are often diagnosed before the age of 40.^1,2 Women are also at high risk of developing lobular carcinoma of the breast. Given the significant risk of developing gastric cancer in their lifetime, prophylactic total gastrectomy is recommended for patients who are found to have a CDH1 mutation on genetic testing.

The patient is a 39-year-old woman with a history of rectal cancer many years ago for which she was successfully treated and had subsequent genetic testing that revealed a germline mutation in the CDH1 gene. 

Her abdominal examination is only significant for well-healed incisions from her prior operation (low anterior resection and diverting ileostomy). 

The patient underwent an upper endoscopy that demonstrated a normal-appearing stomach. Biopsies were taken with cold forceps for histologic examination, which revealed parietal cell hyperplasia and no evidence of malignancy.

Patients with HDGC develop the diffuse type (or poorly cohesive type) of gastric cancer that are submucosal in nature, which means that they are only visible on the mucosal surface when they have violated the mucosa much later in the disease process. As such, prophylactic total gastrectomy is recommended as the only reliable way to prevent a potentially fatal gastric cancer in patients with HDGC.

For patients who are younger than ~20 years of age and for those who decide against operative management, endoscopic surveillance with serial upper endoscopies is an alternative option. The exact frequency of surveillance is unclear, but most guidelines recommend endoscopic exams be performed every 6–12 months, with a minimum of 30 biopsies taken, 5 from each of the following anatomic zones: prepylorus, antrum, transitional zone, body, fundus, and cardia.^3–5 It should be made clear to patients that even with careful endoscopic evaluation, an early gastric cancer can still be missed given that the cancer infiltrates are small and widely distributed.⁶ However, studies suggest that patients who defer gastrectomy achieve similar survival outcomes compared to those who undergo upfront gastrectomy, understanding that most patients eventually opt for prophylactic total gastrectomy after longitudinal counselling.⁷

Though this video demonstrates a prophylactic total gastrectomy via the traditional open approach, many surgeons now perform this procedure in a minimally invasive fashion, either laparoscopically or robotically. Though there are no differences in the rates of serious morbidity or mortality, the minimally invasive approach is associated with less postoperative pain and a correspondingly shorter length of stay than the open approach.

The primary goal of prophylactic total gastrectomy is to remove all gastric mucosa and thereby eliminate the risk of gastric cancer. As such, frozen section pathologic evaluation is performed intraoperatively to ensure that the proximal margin of the gastrectomy specimen demonstrates 100% esophageal squamous mucosa and that the distal margin demonstrates the Brunner’s glands seen only in the duodenum.

For patients undergoing total gastrectomy, we place them in supine position and enter the abdomen via an upper midline incision. The peritoneal surfaces and liver are carefully inspected for metastatic disease. We then take down the falciform ligament before placing the Thompson retractor.

Once the retractor has been placed, we begin by entering the lesser sac through the gastrocolic ligament. We then carry the dissection leftward, dividing the left gastroepiploic vessels and all of the short gastric vessels with a vessel-sealing device (LigaSure). The dissection is then carried rightward along the greater curvature towards the gastrocolic trunk. Here, the right gastroepiploic vein and artery are isolated and ligated with the LigaSure. Once this has been done, the first portion of the duodenum is divided with a single firing of the Endo-GIA 60-mm tan load stapler.

Attention is then directed towards the lesser curvature. We open up the lesser omentum and take the right gastric vein and artery with the LigaSure. We then divide the phrenoesophageal ligament and dissect out the lower 4–5 cm of esophagus within the abdomen and divide both vagus nerves. This provides mobility to the proximal stomach and distal esophagus that, together with division of the duodenum, enables the stomach to be mobilized in the cephalad direction to permit easier dissection of the left gastric vein and artery, which we divide with Endo-GIA tan load staplers (Figure 1).

A site for division of the esophagus is then chosen 1–2 cm above the gastroesophageal junction. We perform this with cautery, placing 3-0 silk traction sutures in 4 quadrants as the esophagus is divided. Once the esophagus has been divided, the stomach specimen is then passed off the field and sent for frozen section examination to ensure that there is no residual gastric mucosa at the proximal and distal margins.

Since the total gastrectomy performed for HDGC is prophylactic, a D1+ or D2 lymphadenectomy is not necessary. However, the immediate perigastric lymph nodes (which constitutes a D1 lymphadenectomy) are removed en bloc with the gastrectomy specimen.

The esophagojejunostomy is performed in an end-to-side, stapled fashion with a 25-mm Covidien EEA stapler. We first place a 3-0 PDS purse string suture in the cut edge of the esophagus to secure the anvil of the EEA stapler. After this, we create our retrocolic Roux limb measuring about 60 cm in length by dividing the jejunum ~30 cm distal to the ligament of Treitz. A two-layer, handsewn, side-to-side jejunojejunostomy is performed with an inner layer of running, full-thickness PDS sutures and outer rows of interrupted 3-0 silk Lembert sutures. We close the mesenteric trap with 3-0 silk sutures.

The Roux limb is passed in retrocolic fashion through a rent in the transverse mesocolon to the left of the middle colic vessels. Be careful not to twist the Roux limb as it is advanced to the esophagus. We then perform the stapled esophagojejunostomy anastomosis by cutting the staple line off the Roux limb and advancing the stapler through the enterotomy. The spike is then advanced out the antimesenteric border of the Roux limb before it is married to the anvil in the esophagus. The stapler is then fired, and the esophageal donut is sent for final margin assessment. Our anesthesia colleagues then advance a nasogastric tube through the anastomosis and into the Roux limb. The overhang on the Roux limb is then excised with a single firing of the Endo-GIA tan load stapler (Figure 2). Finally, we reapproximate the esophageal hiatus with a few interrupted 2-0 silk sutures and secure the Roux limb to the peritoneum of the diaphragm with interrupted 3-0 silk sutures. The mesenteric trap at the transverse mesocolon is also closed with interrupted 3-0 silk sutures.

At experienced referral centers, total gastrectomy can be performed with minimal morbidity and mortality. In a recent retrospective review of 148 patients undergoing total gastrectomy at our institution, the 90-day major morbidity and mortality rates were 14% and 3%, respectively. The median length of stay was 8 days, and the readmission rate was 22%, primarily secondary to dehydration or nutritional compromise.⁸ The patient in this case had an uncomplicated hospital course. She underwent an upper GI swallow study on postoperative day 2, which was negative for an anastomotic leak, and she was discharged to home on a full liquid diet on postoperative day 3.

TOTAL GASTRECTOMY:  

Multifocal adenocarcinoma, poorly cohesive type, confined to the lamina propria (see synoptic report). There is no evidence of malignancy in nine lymph nodes (0/9).

FINAL ESOPHAGUS MARGIN:  

Segment of esophagus with no diagnostic abnormality recognized. There is no evidence of malignancy.

TUMOR STAGE SUMMARY: m p T1a N0.

TUMOR SIZE (greatest dimension): up to 0.1 cm (see comment).

WHO CLASSIFICATION: Poorly cohesive carcinoma (includes signet ring cell carcinoma and variants).

HISTOLOGIC GRADE: G3 (Poorly differentiated).

EXTENT OF INVASION: Tumor invades lamina propria only.

PRIMARY TUMOR: pT1a (Tumor invades lamina propria or muscularis mucosae).

SMALL VESSEL (BLOOD/LYMPHATIC) INVASION: Absent.      

LARGE VESSEL (VENOUS) INVASION: Absent.      

PERINEURAL INVASION: Absent.

PROXIMAL ESOPHAGEAL MARGIN: Squamous lined esophagus, not involved by invasive carcinoma (part B).

DISTAL DUODENAL MARGIN: Not involved by invasive carcinoma.

REGIONAL LYMPH NODES: pN0 (No regional lymph node metastasis): Number of lymph nodes examined: 9.

A self-retraction system such as a Bookwalter, Thompson, or Omni surgical retractor can greatly facilitate exposure throughout the operation. A LigaSure device can facilitate an efficient, hemostatic dissection and allow for division of even major vascular pedicles such as the right gastroepiploic pedicle without the need for ties or staplers. Both the duodenum and the jejunum are divided with an Endo-GIA tan load stapler with tri-staple technology (0.95-1.12 mm staple height), and the esophagojejunostomy is performed with a 25-mm Covidien EEA stapler (4.8 mm staple height).

Nothing to disclose.

The patient referred to in this video article has given her informed consent to be filmed and is aware that information and images will be published online.