Navigating Tough Decisions in Gastric Polyp Management

Recent Guidelines Attempt to Bring Uniformity to an Area of Considerable Practitioner Variation

David A. Johnson, MD


December 15, 2015

This feature requires the newest version of Flash. You can download it here.

Polyps: Commonly Encountered, Frequently Confounding

Hello. I'm Dr David Johnson, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia.

Today, I'd like to discuss the endoscopic management of premalignant and malignant conditions in the stomach. Many of us often encounter stomach polyps of questionable origins in our practices and ask ourselves, or are asked by our patients: What needs to be done regarding possible interventions?

There seems be considerable variation among individual practitioners in responding to this question, which led the American Society of Gastrointestinal Endoscopy (ASGE) to put forth recent guidelines[1] on how to best manage these instances.

Premalignant Conditions of the Stomach

Gastric Polyps

Before we get into the strengths and some potential weaknesses of their recommendations, it's valuable to first provide some important background, beginning with premalignant conditions of the stomach.

Gastric polyps are commonly seen in our practices, most typically in the form of gastric sporadic epithelial polyps. These are often identified by the endoscopist, yet unlike with the colon, nothing we can do endoscopically or with any advanced imaging can inform us as to the polyp's histology.

The majority of gastric epithelial polyps, 70%-90%, are fundic gland polyps, which pose no significant overall risk to the patient in the absence of familial polyposis.

Conversely, although most may assume that hyperplastic polyps would be benign in the colon, we are beginning to understand aspects of these lesions that challenge this idea. Hyperplastic polyps, particularly those located above the sigmoid colon, are associated with an increased risk for gastric cancer. Dysplastic elements are seen in 5%-19% of these polyps. This led to the recommendation to resect these hyperplastic polyps when they measure approximately 5 mm in size. However, it is now almost universally accepted that resection should occur at 1 cm instead. There are some incremental risks that suggest dysplasia, in particular polyps the size of 1 cm or greater and pedunculated morphology. These are some of the things I recommend looking for in patients diagnosed with a hyperplastic polyp in the stomach.

Adenomatous polyps are clearly precancerous and should be endoscopically removed when possible. The recurrence rate is approximately 2.5% when these polyps are completely excised by endoscopic mucosal resection. Therefore, endoscopy is typical a year after the adenoma is removed, with follow-up surveillance at 3- to 5-year intervals thereafter. This is recommended by the current guidelines, even though there is limited supporting evidence.

Patients with Lynch syndrome and familial adenomatous polyposis typically have fundic gland polyps. These are found in close to 90% of children and adults with familial adenomatous polyposis. They do not typically pose a significant risk, although there are usually several of these polyps present, and they are typically difficult to remove. The risk for gastric cancer is incompletely characterized here; as a result, there is no official recommendation in these patients.

Adenomas can also occur in familial adenomatous polyposis. When present, they are usually solitary, sessile, and located typically in the antrum. The relative risk for this syndrome is also somewhat undefined due to its relative rarity. Consequently, the ASGE currently offers no significant recommendations regarding its management.

This is also true of Lynch syndrome patients, where there are very conflicting data regarding the relative risk for gastric cancer. Recently, separate guidelines on Lynch syndrome were issued by the Multi-Society Task Force on Colorectal Cancer[2] and the American Gastroenterological Association (AGA),[3] neither of which offered strong recommendations regarding screening for gastric cancer. Though not uniformly recommended, a one-time endoscopy is thought to be reasonable and justified, particularly in the presence of symptoms.

Gastrointestinal Metaplasia

Gastrointestinal metaplasia also requires our attention, given its relatively frequent diagnosis upon biopsy and the knowledge that it represents 10-fold increased risk for gastric cancer relative to the general population worldwide. However, in the United States, the risk for progression to cancer is low; and, independent of relative risks such as family history of gastric cancer, there seems to be a normative risk. This means we do not always need to intervene once gastrointestinal metaplasia has been identified or even perform routine surveillance. If you do initiate surveillance of these patients, endoscopies can be suspended when 2 in a row have failed to detect dysplasia. If Helicobacter pylori is identified, it is reasonable to eradicate it in the presence of intestinal metaplasia, but again entering such patients into gastrointestinal metaplasia surveillance is likely unnecessary and in line with the ASGE's guidelines when not including other relative risks.

Pernicious Anemia

Pernicious anemia is encountered not infrequently as an endoscopic referral in our practices and has a 3-fold relative risk for atrophic gastritis associated with gastric cancer. However, the incidence of gastric cancer seems to be relatively equal to the normative risk for the population at large. Current recommendations state that these patients should have a one-time endoscopy within 6 months of the diagnosis of pernicious anemia but do not need to be surveyed thereafter once it is clear that they have no evidence of any types of dysplasia.

Gastric Carcinoid Tumors

A concern for many practitioners, gastric carcinoids are best assessed according to their four unique types.

Type 1 is multifocal, well-differentiated, and associated with chronic atrophic gastritis. Type 2 is also multifocal and well-differentiated yet associated with Zollinger-Ellison syndrome or multiple endocrine neoplasia. Type 3 is solitary, well-differentiated, and sporadic and differs from type 1 in its absence of atrophic gastritis. Lastly, type 4 is sporadic, solitary, and poorly differentiated.

Management strategies. The management of gastric carcinoid is critically dependent on the presence or absence of atrophic gastritis. In type 1, which we see most commonly in our clinical practices, data show that the 5- and 10-year survival of these patients is no different from the general population. Therefore, in my opinion, the "high alert" around type 1 does not appear justified ,and the clinical management is poorly defined. Endoscopic surveillance alone can be justified vs polypectomy or endoscopic mucosal resection, but it must be recognized that these are submucosal lesions that need to be individualized based on relative risk.

Patients with type 2 seem to have a significant relative risk for gastric cancer. When these present in Zollinger-Ellison and multiple endocrine neoplasia, they have significant risk for lymph node involvement. Metastases are observed in approximately 10%-30% of lesions upon discovery.

It is important to remember that a major point of differentiation when judging these types is whether or not there is an emphasis of chronic atrophic gastritis, which is something we see in type 1 but not in type 3 or 4. This indicates that types 3 and 4 tend to be much more aggressive. The 5-year survival rate for type 3 gastric carcinoids is unfavorable. Even very small type 3 and 4 lesions should be considered for endoscopic or surgical resection, particularly type 4 gastric carcinoids that have a 1-year survival rate following diagnosis of just 50%.

Postgastric Surgery

Postgastric surgery is also reviewed in the current ASGE guidelines. Although studies have demonstrated some incremental risk 15-20 years after gastric surgery is performed, there is no significant recommendation by this guideline committee that suggests that we should be using different monitoring approaches for such patients. Instead, independent screening choices here should be symptom directed (eg, when iron deficiency is present) and within that 15- to 20-year timeframe.

Malignant Conditions of the Stomach


When it comes to malignant conditions of the stomach, the ASGE guideline committee recommends that if you observe an ulcerogenic lesion, you should take at least seven biopsies of the heaped up edges of the ulcer, something that I regard to be standard of care. Proper diagnosis may be difficult in patients with linitis plastica, which can be missed entirely by biopsies. In such instances, an x-ray, CT, or upper GI may prove helpful. There is additional value in the use of a tunnel biopsy, which I refer to as a three-on-three biopsy. This process is performed during endoscopic biopsy, whereby you re-enter the same incision multiple times and take samples, eventually burrowing your way into the deeper area of the mucosa and submucosa. This may lead to improved histologic evidence to aid proper diagnosis. For all of these strategies, it is recommended that you stage endoscopically using endoscopic ultrasound.

The recommendations have a tremendous void when discussing gastric cancer in that they provide no guidance as to when these patients should be surveyed after surgical resection. My typical practice is to bring these patients back in 6-12 months, then follow them on an annualized basis for 3 years, and finally to somewhat arbitrarily follow them on a 3- to 5-year interval thereafter. I should note that there is no evidence that I can cite to support that recommendation; it is simply my preferred practice.

Mucosa-Associated Lymphoid Tissue Tumors

Mucosa-associated lymphoid tissue (MALT) tumors are low-grade lymphomas known as a extranodal marginal zone B-cell lymphomas. MALT tumors are associated in particular with the infection H pylori. We often see these patients in our practices when they come for evaluation of dyspepsia, weight loss, and sometimes GI bleeding. When in an early stage of lymphoma, the vast majority of these patients (around 80%) respond to directed therapy for H pylori.

In the past, the optimal surveillance recommendation for these patients was unclear once MALT lymphoma was established. Current recommendations state that sampling should occur every 3-6 months for the first 2 years, which should be extended to every 6-12 months after that for approximately 3.5 years. Endoscopic ultrasound can be very helpful in the evaluation of these patients and provides a glimpse at lymph node involvement, which could positively affect staging criteria.

GI Stromal Tumors

GI stromal tumors (GISTs) are something that we also encounter frequently in our practices, being the most common type of mesenchymal stomach tumors. These tumors are classified by the National Institutes of Health guidelines, using size and mitotic index for high-powered field. However, GISTs require surgical resection and do not allow the use of endoscopic biopsies to predict relative risk.

Generally, the guidelines suggest that any symptomatic lesion be surgically removed, especially if the lesion is a source of bleeding and/or is larger than 2 cm. Endoscopic ultrasound should be considered upon any suggestion of malignant features with these lesions. If they are less than 2 cm, the guideline committee suggests that annual surveillance for GISTs less than 2 cm is optimal. However, it should be noted that the evidence on this subject is somewhat weak.

Judging the Strongest and Weakest Evidence

There are four areas where the guideline committee felt that the level of evidence was particularly compelling, which in turn led to their strongest recommendations.

Firstly, the committee recommended that gastric polyps undergo biopsy and resection when possible. Secondly, in the setting of multiple polyps, they recommended that endoscopic resection of the largest polyps should be done followed by representative sampling of the other polyps to guide the potential intervention. Thirdly, it was recommended that with potentially malignant ulcers, seven or more biopsies should be obtained from the heaped up hedges, particularly to exclude for gastric cancer. Lastly, and somewhat obviously, it is recommended that endoscopic ultrasound with or without fine needle aspirate be performed in all gastric submucosal lesions.

Conversely, there are notable limitations to their other recommendations that are supported by less evidence. Polypectomy of fundic gland polyps 1 cm or larger, hyperplastic polyps 5 mm or larger, and of course all adenomatous polyps irrespective of size has been recommended, though the evidence base around this is relatively weak. In the setting of multiple hyperplastic or even adenomatous polyps, systematic sampling of the area was recommended to assess for H pylori and metaplastic atrophic gastritis.

Surveillance endoscopy and gastrointestinal metaplasia should probably be reserved for patients who have incremental risk. In the United States, intestinal metaplasia is not in and of itself considered to be beyond the normative risk in the absence of family history or age. Here I would recommend taking a deep breath, stepping back, and following these patients in a more normative symptom-directed way. In endoscopy and the diagnosis of pernicious anemia, the choice would be a one-time endoscopy within 6 months of diagnosis. These patients do not need to be surveyed thereafter in the absence of upper GI symptoms.

Finally, GIST tumors smaller than 2 cm can be endoscopically followed on an annual basis if surgical resection is not performed. This is considered a relatively weak recommendation based on the evidence but remains the standard we follow in our practice.

Hopefully, this overview has given you insight into the common problems we deal with regarding when to initiate and stop surveillance in certain patients, as well as when to be alarmed about premalignant and malignant conditions of the gastric body. It is my hope that it will provide meaningful input for your forthcoming patient discussions.

I am Dr David Johnson. Thanks for listening. See you next time for another GI Common Concerns—Computer Consult .


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: