When Should Endoscopy Be Done in the Patient With Reflux?

M. Brian Fennerty, MD


March 10, 2003

In This Article


Historically, subspecialists have often depended on objective diagnostic tests in the initial management of disease processes (ie, gastroenterologists would use a test such as endoscopy to diagnose gastroesophageal reflux disease [GERD]), whereas primary care physicians (PCPs) have often relied on clinical findings or the results of trials of empirical therapy to make a disease diagnosis. This disparity in initial clinical approach in GERD management between specialist gastroenterologists and PCPs is now especially questionable, given that community-based studies have shown that 40% to 60% of patients with reflux symptoms have no objective endoscopic findings of GERD.[1] Thus, endoscopy only has a sensitivity of approximately 50% for diagnosing GERD. It is my opinion that this relatively modest (at best) sensitivity for endoscopy as a diagnostic test for GERD is underappreciated by many PCPs referring patients with suspected GERD for this procedure as a diagnostic tool. Furthermore, the charges for upper gastrointestinal tract endoscopy and the additional hidden costs (lost productivity, wages, etc, for the patient on the day of the procedure), result in a substantial societal and healthcare system burden financially when endoscopy is used in this setting as a routine diagnostic test. Thus, referring every patient with reflux symptoms for endoscopy is not justifiable, given the substantial cost coupled with the poor sensitivity.

By contrast, given the effectiveness of potent antisecretory agents (ie, proton pump inhibitors [PPIs]) in relieving reflux symptoms and diagnosing GERD, a trial of PPI therapy is more cost-effective than endoscopy for confirming a diagnosis of GERD in an otherwise uncomplicated case.[2] It is my recommendation that referral for endoscopy in a patient with suspected GERD should be reserved for patients with symptoms of GERD who also have alarm symptoms (discussed below), those patients who do not respond to a trial of empirical therapy with a PPI, or those in whom the diagnosis of GERD remains uncertain after initial assessment. Additionally, endoscopy may also have value in the patient with well-established and well-controlled GERD symptoms as a screening exam for Barrett's esophagus (a premalignant condition in which the normal squamous epithelium of the esophagus is replaced in varying degree by a metaplastic specialized columnar epithelium).[3] However this latter use of endoscopy in a patient with GERD is quite different in that it is done in the setting of chronic disease and for a very different purpose than for making a diagnosis of GERD. Furthermore, the recommendation to screen all patients with chronic GERD symptoms for Barrett's esophagus remains controversial.[4]