Surveillance Is Almost Never Appropriate

As physicians, we have grown accustomed to the concept of screening for and surveillance of those found to have premalignant conditions with the expectation that by doing so we can prevent cancer and do little if any harm. In reality this concept has rarely been demonstrated to be true and indeed we often do great harm. Examples of this fallacy include the widespread use of prostate-specific antigen (PSA) to screen for prostate cancer (resulting in us resecting many “harmless cancers” while leaving men impotent), mammography starting at age 40 to screen for breast cancer (finding harmless cervical intraepithelial neoplasia lesions resulting in disfiguring surgeries), and in our own field of gastroenterology in screening patients with chronic gastroesophageal reflux disease (GERD) and finding short segment non-dysplastic Barrett’s esophagus (BE) and enrolling those patients in an endoscopic surveillance program (leading to expensive endoscopies, cancer phobia and no evidence that this improves cancer outcomes). So now we as a gastroenterology community are discussing surveillance of a ubiquitous and even lower risk lesion — gastric intestinal metaplasia or GIM — than short segment non-dysplastic BE? Come on, you cannot be serious! Have we learned nothing from the above examples?1

Let’s look at the facts. Gastric cancer is not a common cancer in the Western world or the U.S. (1.7 percent of all cancers) and not all gastric cancers proceed through the intestinal metaplasia (IM) pathway. Furthermore, GIM is not the premalignant lesion, dysplasia in a field of GIM is the premalignant lesion. Twenty-five years ago, we demonstrated that in the U.S., approximately 15 percent of Caucasian adults at very low-risk for gastric cancer have GIM, and as many as 50 percent higher but still low-risk men (Blacks, Latinos, Asians) have GIM.2 Thus a conservative estimate is that of the about 250 million adults in the U.S., somewhere around 50 million harbor GIM. Yet there were only 28,000 cases of gastric cancer this year (and again not all linked to the GIM pathway). Assuming 50 percent of these cancers proceed through a GIM pathway, the incidence of cancer in someone with GIM in the U.S. is approximately 0.00011. Nearly 8,750 people would have to be screened to detect one case of gastric cancer. So, should we survey everyone with GIM? You’re kidding, right?

So, where does the concept that we should be surveying these folks come from? Not from ASGE — their position is to perform “surveillance endoscopy for patients with GIM who are at increased risk for gastric cancer due to ethnic background e.g. (East Asia, Latin America mountainous regions) or family history. Optimal surveillance intervals have not been extensively studied and should be individualized.”3 What ASGE is really saying is unless there is a very compelling reason to survey, don’t do it. ASGE also identifies an additional group who should be surveyed and that is those with GIM and dysplasia. But these above “survey eligible” populations represent only a small percentage of all our patients with GIM.

Nearly 8,750 people would have to be screened to detect one case of gastric cancer.

Surprisingly, many of the European societies are much more liberal in their potential GIM surveillance candidates.4 While they wisely recommend not surveying those with GIM isolated to the antrum, they unwisely recommend surveillance for all of the following: those with more extensive GIM (two or more areas of the stomach); those with the incomplete GIM subtype; those with a family history of gastric cancer; those of a non-Caucasian race including Blacks, Asians and Latinos; and first-generation immigrants from areas of high gastric cancer incidence. Following these recommendations would mean over 50 million U.S. adults could be candidates for GIM surveillance. This statement even makes PSA screening in all men sound rationale. This approach to nearly universal surveillance of GIM is neither scientific nor practical, much less cost-effective or safe.

What then should we do when we encounter a patient who has a biopsy from their stomach that is read by the pathologist as having GIM? First, we should reassure the patient they have a common finding and they have little if any risk for cancer. We should assess if there was dysplasia, whether the IM was widespread, where they were born and whether they have any first-degree relatives with gastric cancer. If they have isolated IM, no dysplasia, were born in a low-risk region for gastric cancer (e.g. Europe and the U.S.) and no family history of gastric cancer, my recommendation is straightforward and simple: no surveillance. Even in those with more extensive GIM I believe this to be the correct recommendation: no surveillance.

I would recommend surveillance only in those very few individuals with GIM associated dysplasia, those with GIM and first-degree relatives with gastric cancer, or those first-generation immigrants from a high gastric cancer incident region that has extensive GIM. Optimal surveillance intervals for these few surveillance candidates are unknown, but for now I would recommend at least yearly surveillance in those with dysplasia (after complete resection of their dysplasia) and surveillance of every two to three years in these few other higher-risk GIM individuals.

Remember; above all: first, do no harm.

Ernst J. Kuipers, MD, PhD, AGAF, provides a different view on surveillance.

Dr. Fennerty has no conflicts to disclose.

1. Fennerty, M.B.Gastric intestinal metaplasia on routine endoscopic biopsy. Gastroenterology. 2003;125:586–590.
2. Fennerty, M.B., Emerson, J.C., Sampliner, R.E., McGhee, D.L., Hixson, L.J., Garewal, H.S. Gastric intestinal metaplasia in ethnic groups in the Southwestern United States. Cancer Epidemiol Biomarkers Prev. 1992;1:293–296.
3. Evans, J.A., Chandrasekhara, V., Chathadi, K.V. et al, The role of endoscopy in the management of premalignant and malignant conditions of the stomach. Gastrointest Endosc. 2015;82:1-8.
4. Dinis-Ribeiro, M., Areia, M., de Vries, A.C. et al, Management of precancerous conditions and lesions in the stomach (MAPS): guidelines from the European Society of Gastrointestinal Endoscopy (ESGE), European Helicobacter Study Group (EHSG), European Society of Pathology (ESP) and the Sociedade Portuguesa de Endoscopia Digestiva (SPED). Endoscopy. 2012;44:74-94


  • This is an intriguing article (and reality check of sorts) in the debate regarding intestinal metaplasia of the stomach and what to do with respect to surveillance. It is a must-read to help contextualize this clinical dilemma. However, a few things would have made this more pleasant and robust:
    1. The punctuation is abysmal. Even my elementary schoolchildren were able to point out the numerous missing commas, e.g. between two independent clauses. Run-ons are do not inspire trust.
    2. The verb “to survey” is not the root of the word “surveillance”. A quick glance at an English language dictionary, e.g., would quickly clarify this. “Survey” is not a conjugation of “surveil”.

    Without intending to be overly critical, we should safeguard our field and ensure that biomedical writing not be exempt from the basic rules of the English language.

    • There is an error in the sentence “Run-ons are do not inspire trust.” (the word “do” should be omitted). Otherwise, spot on with the comment! And may the verb “to surveil” live on!!

Join the discussion

Your email address will not be published. Required fields are marked *