Invited Commentary: Performing and Reporting Upper Endoscopy: Follow the Surgeon’s Lead

Upper endoscopy is for all practical purposes a foregut gastroenterologist’s physical exam. It is the first and most important diagnostic study I perform if it has not been done prior to me seeing the patient. It is the most important report I look to review if the patient is referred or comes for a second opinion. I look for a report quite similar to what Louie and Padhya ¹ have described in their excellent article in this issue, addressing the crucial components of an upper endoscopy.

My approach to endoscopy is simple. My intent is to make a diagnosis. I begin if I am able with an exam of the larynx, especially in those with LPR complaints, and make every effort to do a careful direct intubation of the esophagus. That gives me an impression of upper sphincter tone. I teach GI fellows that the esophagus is not a simply a conduit to the stomach, emphasizing a careful look going both in and on withdrawal. I look for signs suggestive of eosinophilic esophagitis and include an endoscopic reference score (EREFS) ² if my findings suggest a clinical suspicion. I obtain my distal biopsy approximately 3 cm above the squamocolumnar junction (SCJ) and “proximal” about 15 cm above that. Five biopsies are optimal and should be at different levels to account for the patchy nature of the disease. ³ My esophageal photographs are similar to those suggested by Louie and Padhya though I vary my esophageal body pictures to visible abnormalities and if there is no proximal finding (inlet patch, diverticulum) I might not always photograph the very proximal esophagus. After reading their article I plan to make every effort to include this. I must say that I do not regularly record pictures of the normal stomach other than a retroflexed view of the GEJ and only take pictures of the duodenum to document that I reached the second portion, unless the findings require photographic documentation. A clear recording of the distance of the Z line from the incisors, the LA grade of esophagitis, a description of the Hill grade and the axial length (size) of the hiatal hernia in centimeters (or better the level of the diaphragmatic pinch, top of gastric folds, and squamocolumnar junction) with accompanying photos allow me to “remember what I saw” when the patient comes for follow up and allows my foregut surgical colleague to “see” what I was looking at. A description of the mucosa when eosinophilic esophagitis is suspected and an EREFS score support my pushing the pathologist to provide more information regarding eosinophil counts or other biopsy findings if I need them. The presence of an inlet patch, though felt to be of little clinical importance is not always an incidental finding as it may be secretory ⁴ and perhaps contribute to LPR symptoms, with an occasional patient responding to RFA ablation ⁵ and add to my understanding of the patient’s symptoms. Many a patient with LPR symptoms is reassured when I tell them the larynx is normal. My retroflexed view of the cardia (not always a simple task with modern endoscopes) includes 30 to 60 seconds of observation of the valve with continuous CO₂ insufflation. A quick look might underestimate a Hill grade 3 as a Hill grade 1; crucial as a grade 3 valve and/or hiatal hernia greater than 2 cm usually takes a primary TIF off the table. A Hill grade 3 or 4 heightens my suspicion of GERD. Hernia’s greater than 4 cm are more likely to need repair for symptoms, and are an overlooked source of GI bleeding even in the absence of Cameron ⁶ lesions. My foregut surgical colleagues appreciate the description and photographs of the GEJ when assessing the patient for possible surgical intervention. A clear description of any columnar lined esophagus using the Prague criteria are routine. I use narrow band imaging regularly in making decisions regarding biopsy of segments of suspected Barrett’s <1 cm. I tend to prefer the term variable to irregular in describing the SCJ when it is not sharp. I expect my report to be able to present a clear description (in words and photographs) of the esophagus, stomach, and duodenum.

When the “outside report” lacks the components or photographs outlined in my surgical colleagues’ superb article I invariably repeat the exam, as I cannot be assured a careful and thorough exam have been done. I respect that most endoscopists know what is “normal” but there are enough repeat endoscopies with Hill valve grade 3 or 4 that has no notation in the report, a hiatal hernia described as small or medium that is >4 cm, or has a para esophageal component or an incomplete description of a previous fundoplication that I cannot always be confident that the first endoscopist is looking for the same things I am. I understand that reports take time, that we sometimes don’t feel like we have, so a normal esophagus, normal stomach, etc. is easy to “click” in the endo writer. Good, clear photographs are not always easy to take. Adjudicating a blinded multicenter erosive esophagitis trial makes that quite evident. Writing a report in anticipation of the patient going elsewhere or needing a consult is I suspect rarely thought of at the index endoscopy.

I do often ask myself if my “thoroughness” is over the top and if I really need the extra minute to include the components of an endoscopy suggested by Louie and Padhya in every report or if the effort to do a careful retroflexed view of the cardia and the region of the GEJ and the additional 30 seconds to a minute is to the patients benefit. My strong bias is that it is and can be done with little sacrifice of daily volume, even in a busy practice. A minute or 2 of additional recording time at the most is all that is required using the currently available endo writers. Presenting a narrative both in words and photos benefits both endoscopist, consultant and the patient.

I have not used a template like the one used by Dr Louie’s group, nor have we at our institution looked carefully at consistency of upper endoscopic reports, as we do with adenoma detection rate. Perhaps for good reason as data on the importance of the findings discussed in their article on clinical outcomes are lacking. After reading this article I suggest we as physicians dedicated to care of patients with foregut disease consider the importance of doing so and begin collecting the data. I will continue to teach my approach to upper endoscopy with trainees and encourage all AFS members, especially my gastroenterology colleagues to adopt a similar approach. In the interim I will follow my surgical colleague’s lead.