dyspepsia: endoscopy usually unnecessary

 An updated systematic review/meta-analysis reaffirmed the general lack of utility of endoscopy in patients who have symptomatic dyspepsia but without alarm features (see dyspepsia review of endoscopy findings ClinGastroHep2022 in dropbox, or doi.org/10.1016/j.cgh.2022.05.041)

 

Details:

-- 15 cross-sectional surveys, including 6 printed from 2010 until January 2022, analyzed the prevalence of dyspepsia as well as endoscopy (EGD)

-- EGD was performed in all participating individuals regardless of their symptoms in 6 studies, in all individuals with dyspepsia in 2 studies, and in some patients having EGD regardless of symptoms in the remaining 7 studies

-- 41,763 people underwent EGD, of whom 16,900 (40.4%) had dyspepsia

    -- the prevalence of dyspepsia in these surveys range from 2.4% to 68.5% with a pooled prevalence of 26.6% (13.6%-42.1%), but with lots of heterogeneity between the studies

-- 5 studies were conducted in the West, and 10 in the East (prevalence of gastroesophageal cancer is higher in the East)

-- there were differing criteria used to define dyspepsia in the studies, with a broad definition in 5 studies, the Rome III  criteria in 6 studies, the Rome II criteria in 3 studies, and the Rome I criteria in the remaining study

-- 9 studies defined dyspepsia as symptom duration of 3 months or less and 4 studies had range of 6 to 12 months; both ranges had pooled prevalence of dyspepsia of 30%

-- 2 of the new papers added to the systematic review were population-based, leading to a total of seven population-based studies

 

Results:

-- erosive esophagitis (reported in 12 studies with 3435 subjects having dyspepsia):

    -- found in 12.0% (3.6%-24.2%), with significant heterogeneity between the studies

-- Barrett’s esophagus (reported in 10 studies with 3004 people having dyspepsia):

    -- found in 0.6% (00.0%-1.9%), also with significant heterogeneity

-- gastroesophageal cancer (reported in 11 studies with 3350 subjects):

    -- 4 esophageal cancers were found in those with dyspepsia who underwent EGD, 7 gastric cancers (pooled incidence of both: 0.32%)

-- peptic ulcer disease (15 studies with 3965 subjects having dyspepsia):

    -- pooled prevalence 7.2% (4.0%-11.1%), with statistically significant heterogeneity

    -- 11 studies separated gastric and duodenal ulcers: with pooled prevalence of 2.7% (1.4%-4.3%) and 3.3% (1.1%-14%), respectively

-- specific assessment of the two newly added papers that were population-based had very similar findings for peptic ulcer (9.3%); gastric ulcer (2.5%); duodenal ulcer (4.9%), and erosive esophagitis (14%)

 

-- studies reporting clinically significant endoscopic findings independent of dyspepsia symptoms

    --10 studies reported the prevalence of endoscopic findings in 40,084 individuals regardless of symptom status, finding:

       -- erosive esophagitis: still the most common, 11% (8.9%-13.2%)

       -- peptic ulcer disease: 4.4% (2.5%-6.7%)

    -- 8 studies reported Barrett’s esophagus, with a pooled prevalence of only 0.3% (0%-1.2%)

    -- 9 studies with 39,062 participants detected 18 esophageal cancers and 122 gastric cancers with a pooled prevalence of 0.09%, with statistically significant heterogeneity

 

-- comparing EGDs in those with dyspepsia symptoms versus those without symptoms:

    -- overall, more than 85% of EGDs were completely normal

    -- the only statistically significant difference was in those with peptic ulcer disease, OR 1.61 (1.08-2.39), which was only significant for duodenal ulcer

    -- no statistically significant difference between Barrett’s esophagus (OR  1.08 (0.61-1.89)) or erosive esophagitis (OR  0.89 (0.76-1.05)) in those symptomatic or not

    -- gastroesophageal cancer was rare (< 0.4%) and equally prevalent among those with and without dyspepsia 

    -- and, the overall prevalence of the composite of any clinically significant endoscopic finding was no higher among those with dyspepsia versus without 

    -- assessing only the population-based studies, the OR for erosive esophagitis was 1.18 (0.89-1.56), not statistically significant, though only 4 studies provided data; peptic ulcer disease remained statistically significantly higher in those with dyspepsia (OR 1.87), again for duodenal ulcer only

 

-- erosive esophagitis was more prevalent in Western studies (25%) versus Asian studies (6.6%); and normal EGDs more common in the East (85.3%) versus the West (66.7%)

  

Commentary:

-- dyspepsia is remarkably common, with a pooled prevalence of about 20% globally, and is associated with decreased quality of life, anxiety about serious GI conditions, and, not infrequently, pleas by patients to have endoscopy to see what is going on

-- the current systematic review/meta-analysis is an update of the 2010 study, adding 6 new studies published since 2010, but all from Asia; this new systematic review included many more EGDs (41,763 in the study versus 5389 in the previous review)

-- it is more useful to us in primary care that they focused on studies that were community-based. These patients are much more like the ones we see, presumably having less likelihood of serious disease (alarm features or being refractory to therapy) than those referred to GI clinics

-- It was pretty striking in this large study that EGD was not particularly helpful in distinguishing those with dyspepsia who had Barrett’s esophagus, erosive esophagitis, or gastroesophageal cancer (they found no difference in the incidence of these whether patients were symptomatic or not). It is notable however that there were very few cases of gastroesophageal cancer in these studies, making overall generalization difficult about the relative risk if symptomatic or not. However, the results of the study would not apply to those who have alarm features associated with dyspepsia, including unintentional weight loss, dysphagia, odynophagia, unexplained iron deficiency anemia, persistent vomiting, palpable mass or lymphadenopathy, or a family history of gastroesophageal cancer; all of these would increase the likelihood of cancer. Additionally, the results of the study may not apply to those who were refractory to medical therapy (again, likely higher risk of cancer or other more severe pathology)

-- this study reaffirms that symptomatic dyspepsia is rarely associated with major adverse upper GI outcomes. Of note, the use of EGD is much more common in the East, where gastroesophageal cancer is much more common, and their guidelines for EGD are much more aggressive than in the West

    -- in fact, one of the findings in the study which seems counterintuitive was that the incidence of erosive esophagitis and Barrett’s esophagus were less common in the studies from the East versus the West, and normal EGDs were found more commonly  in the East (85.3%) versus the West (66.7%)

 

Limitations:

-- no data on the presence of H. pylori. This is a major cause of dyspepsia/peptic ulcer disease/gastric cancer, and has an overall global prevalence of >50% of the population. The prevalence does vary dramatically between Africa (70% of the population) down to Oceania (24%), with the US having a point estimate of 37% (see h pylori global prevalence Gastro2017 in dropbox, or doi: 10.1053/j.gastro.2017.04.022). H Pylori has made it onto the list of carcinogens: see http://gmodestmedblogs.blogspot.com/2022/03/carcinogen-update-now-including-h-pylori.html )

-- unclear why there were more abnormal EGDs in the West than in the East, given that gastroesophageal malignancies are more common in the East. This raises questions about the methodologies in the studies (eg were those patients studied in the West more likely at baseline to have abnormal EGDs than those in the East, where there is a much lower bar to doing EGDs and even done as routine screenings: ie a selection bias). The overall diagnoses of cancer were very low, and notably did not change from earlier review (0.25%) vs this current on e (0.36%)

    -- unfortunately, all of the 6 new studies published since 2010 were from Asia, limiting generalizability of the conclusions to other areas of the world; the results may also be distorted by the huge increase in the numbers of endoscopies done there in those reported studies after 2010 (the "elephant in the room"). In addition, the definitions of dyspepsia were somewhat different in different areas, though the Eastern studies used the more specific Rome criteria

    -- that being said, we do not have enough information about how the new Rome IV  criteria for dyspepsia diagnosis would fare: would these new criteria defining symptomatic dyspepsia result in significant differences in the specificity for erosive esophagitis, for example, enough so  to change the results?

-- this study did find no statistical difference in the important endoscopic findings whether people has dyspeptic symptoms or not.  But the numbers of people with Barrett's esophagus, erosive esophagitis and cancer were very low, rendering comparative statistical analysis pretty meaningless

    -- that being said, these low numbers still reflect the low likelihood of these potentially serious complications in those with dyspepsia, reinforcing the conclusion that endoscopy is not so useful

    -- but, in terms of cancers, these can be really serious (ie fatal, even after major surgery and other treatments), so statistical analysis finding no detection difference in those with vs without dyspepsia symptoms may not be so useful. so, should we just ignore the cancer detection? what if we do not do surveillance EGD: would detection of cancer be equally treatable if we wait to see that there was decreased response to antacid therapies that leads to an EGD? and, some cancers may well respond symptomatically to continued antacid therapies: will we miss them entirely and detect the cancer only after it has spread???

 --in addition, the assessment of dyspepsia symptoms may not be so accurate: some symptoms come and go (so, the presence of “symptoms” may depend on if the individual happened to have potentially transient symptoms at the time of the survey), or baseline psychological issues (some people are more somatic than others; and some are deniers), or even if the patient was just highly stressed out (stress can increase stomach acid production and dyspepsia: eg see https://journals.lww.com/eurojgh/fulltext/2017/05000/perceived_stress_as_a_risk_factor_for_dyspepsia__a.11.aspx )

-- as with any large systematic review/meta-analysis, they are combining the results of very disparate studies with different inclusion/exclusion criteria, methodologies, and overall patient selection (for example, it is not clear that patients with alarm symptoms were excluded in the individual studies); there was no large, really high-quality study which randomized people who were vs weren't symptomatic to EGD and assessed symptoms by Rome IV, for example, and then assessed the correlation between symptoms and findings.

    -- i am also not aware of studies assessing adverse outcomes from Barrett's or erosive esophagitis stratified by symptoms: perhaps those without symptoms have a different progression to cancer??? perhaps symptoms are different in those with low- vs high-grade dysplasia on biopsy?

 

So, an interesting study reinforcing prior studies that EGD had little benefit in those with dyspepsia. This study found, rather strikingly, that there was no difference in pretty much any of the important diagnoses (eg, esophageal/gastric cancer, Barrett’s esophagus, erosive esophagitis) in those with or without dyspeptic symptoms. It reinforced the very low likelihood of cancer. This all suggests several conclusions (with the potential concerns outlined above in "Limitations"):

-- we should dissuade patients with simple dyspepsia (eg no alarm symptoms) that empirical therapy is the appropriate initial approach, and that endoscopy is invasive and occasionally associated with major adverse events

-- there are different approaches to empirical therapy: one is to try the minimalist therapy (start with antacids, then increase systematically to H2 blockers and then to PPIs), with the goal of minimizing the intensity and potential adverse effects of the more potent therapies, but with the caveat to the patient that symptom improvement may take longer) vs going straight to PPIs (with the goal of getting immediate relief, using the response to PPIs as a possible marker of less severe disease, but then trying to step-down to the least potent therapy that works (which, unfortunately, seems to be done pretty rarely: eg see http://gmodestmedblogs.blogspot.com/2022/07/ppis-way-overprescribed.html )

-- and reserving endoscopy for those not responding appropriately to acid suppression, as a marker of higher likelihood of severe underlying disease

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs (2 options):

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list