GERD Consensus 1997

The following algorithm was developed at a meeting (Nov. 15-16, 1997) held at Yale University, School of Medicine, New Haven, and designed to establish a consensus for the treatment of Gastro-Esophageal Reflux Disease (GERD).

In developing the algorithm, it was appreciated that most patients with GERD present to and are managed by general practitioners, some of the care for patients with GERD will require referral to a gastroenterologist and a surgeon, and that the management of GERD was a continuum, with management based upon not only symptoms, but referral patterns and, more increasingly, financial constraints. The first step in the management of the patient with suspected gastroesophageal reflux disease is a history and physical examination. A history of retrosternal burning-type chest pain radiating up from the epigastrium towards the throat is characteristic of GERD, and as emphasized by a recent report, such symptoms have a high specificity for the diagnosis. If the patient has either atypical symptoms, or any alarm symptoms such as dysphagia, anemia, weight loss, predominantly abdominal pain, or pain that does not respond to antacids, or first develops symptoms after the age of 50, then prompt referral for endoscopy is necessary. The majority of patients probably do not require endoscopy initially and such patients should be counselled in regard to more simple anti-reflux measures, which may include: elevation of the head of the bed, weight loss, and the avoidance of precipitating foods and drugs. It is of note, however, that there exists little rigorous data to support the concept that such measures are either of proven or long-term benefit in the management of GERD. The patient should also be advised to take antacids at the time of the symptoms, and to avoid heavy meals late at night or within 3 hours of retiring to bed. If symptoms persist after 2-4 weeks of these simple measures or if antacids are needed often if not daily then formal therapy should be instituted. In the past the first line of treatment was considered to be the use of an H2-receptor antagonist but the current almost uniform positive clinical experience with the use of proton pump inhibitors at the standard (full/low) dose as initial therapy suggests that this strategy is more acceptable to both clinicians and patients. In circumstances where the pain is a consistent component of the symptomatology and particularly if there is a nocturnal component the use of a proton pump inhibitor should be given unequivocal consideration. If the symptoms are more suggestive of a motility problem, a prokinetic drug such as the 5-HT4-receptor agonist cisapride may be included. These drugs should be given initially as a one-month trial, and then discontinued. If symptoms recur after this 4-week trial or if they are not relieved during this treatment period, then the patient should be referred for endoscopy. Although the decision to undertake endoscopy at this time is somewhat arbitrary it is consistent with the widely accepted position that a "once in a life time endoscopy" in reflux disease is acceptable. Thus an endoscopy evaluation can be included in the database of the patient before major (and possibly) long-term therapeutic decisions are undertaken.

At the time of endoscopy, particular attention should be paid to the presence of erythema, erosions, and other signs of esophagitis and, the gastroesophageal junction should be examined carefully. If there is any suspicion of Barrett's esophagus, then four-quadrant biopsies should be taken at the GE junction, but a GE junction of normal appearance should not be biopsied routinely. It was felt by the panel that antrum and corpus biopsies should be taken for the diagnosis of H. pylori even in the absence of peptic ulceration given the recent reports of accelerated gastric atrophy in patients with H. pylori taking long-term proton pump inhibitors. Thus, prophylactic eradication of H. pylori in this group would be mandatory. A minority opinion held that since H. pylori infection is common, and in most circumstances does not lead to harm, establishing whether or not H. pylori was present was irrelevant and potentially confusing.

There was uniform agreement that critical issues in the management of GERD were the need to ensure appropriate effective therapy to minimise the risk of the development of Barrett's esophagus and to identify it accurately. In patients found to have Barrett's esophagus at biopsy but with no dysplasia, the consensus was that such patients ought to be in a surveillance program with approximately 5-year intervals between surveillance endoscopies. This opinion was based upon the cost-effectiveness analysis of Provenzale and colleagues. If the pathologist diagnoses Barrett's esophagus with the presence of dysplasia, then the material should be reviewed by two expert independent pathologists to confirm the diagnosis and to establish whether the changes conform to a pattern consistent with low- or highgrade dysplasia. The identification of low-grade dysplasia mandates repeat biopsy at one year to determine whether there is progression to high-grade dysplasia. If at the one year time point multiple biopsies fail to identify evidence of dysplasia, the patient should return to the normal surveillance program; if low-grade dysplasia persists, then endoscopy and biopsy should again be repeated in one year. If there is a diagnosis of high-grade dysplasia at the initial or any follow-up biopsy, then the patient should be treated with a high dose proton pump inhibitor for three months, e.g. pantoprazole 40 mg b.i.d., omeprazole 20-40 mg b.i.d. or lansoprazole 30 mg b.i.d. for three months, to rule out the possibility of inflammatory pseudodysplasia. If the repeat biopsy after three months confirms high-grade dysplasia then the patient should be referred for consideration of surgery. Some of the experts on the panel felt that very careful repeated endoscopic surveillance of high-grade dysplasia could be performed to monitor for and biopsy any visible mucosal abnormality. Under such circumstances surgery might be obviated until a carcinoma was discovered. However, the majority were uncomfortable with this approach and indicated a preference for prophylactic surgery once the presence of a high-grade lesion had been reliably identified. Ablation therapy by laser, bicap, heaterprobe or photodynamic therapy for Barrett's esophagus and/or dysplasia should be currently regarded as experimental.

In the absence of Barrett's esophagus or for Barrett's esophagus without high-grade dysplasia, the primary goal of treatment of GERD should be to relieve symptoms. Thus with the identification of minimal or modest esophagitis at endoscopy and mild symptoms, it was felt that a proton pump inhibitor should be the initial choice. Little support was expressed for initial treatment with either an H2-receptor antagonist or a prokinetic agent. There was uniform agreement that in the event of moderate or severe esophagitis or particularly troublesome symptoms, all patients should be treated with a proton pump inhibitor immediately. If a patient had been placed on an H2-receptor antagonist or prokinetic therapy and failed to resolve their symptomatology or relapsed it was agreed that a proton pump inhibitor should be prescribed. Under such circumstances it was accepted that proton pump inhibitors such as pantoprazole (40 mg daily) or lansoprazole (30 mg) should initially be given at standard doses. It was noted that 20 mg of omeprazole daily might not provide symptom relief for all patients, and that the dose of this drug might need to be increased to 20 mg b.i.d. or even more, to achieve optimal dosing. In some patients a further increase in dose may be necessary, often requiring a change to b.i.d. dosage. Based upon currently available information it was felt that the addition of a prokinetic agent to therapy in conjunction with a proton pump inhibitor was likely to be of only marginal benefit for GERD. Once a dose of either a H2-receptor antagonist, prokinetic agent, and/or proton pump inhibitor which relieves symptoms is identified, this dose should be maintained for a period of three months. After this time, an attempt should be made to reduce the dose, with the aim of maintaining a stable clinical status (asymptomatic) on half dose PPIs or alternatively on alternative day therapy. If symptoms recur, then the patient should go back to the full dose of proton pump inhibitor and a plan formulated for long-term treatment. The long-term treatment options are either medical therapy, with attempts to reduce the dose of medication occasionally, or consideration for surgery. As regards patients who might enter long term therapy it was noted that many, particularly those in the older age group, may be taking several medications. Thus careful attention should be directed to the drug interaction profile of the various types of proton pump inhibitors.

Surgery is an attractive option for younger patients (age <40 years) and for patients who do not want to take chronic medication for lifestyle reasons, particularly if they have required proton pump inhibitor therapy continually for a year, since they are then unlikely to be able to manage without these medications. Prior to surgery, patients should have the opportunity of a full and frank discussion with both their gastroenterologist and the surgeon, preferably together, regarding all the risks and benefits of long-term medical versus surgical therapy. Although there are no long-term data available, laparoscopic fundoplication generally appears to be safe, and preferable to open surgery. Preoperative motility studies are essential both to precisely delineate the state of the sphincter as well as to facilitate follow-up if there is any post surgical relapse. It was felt by all that the critical determinant of successful surgery was the choice of surgeon and that surgery for gastroesophageal reflux should be confined to those surgeons experienced and expert in surgery for this condition. Furthermore, it was noted that current information supported a "wrap" procedure and that no data existed to suggest that the addition of any form of vagotomy conferred any further benefit. However, above the age of 60, the risks from surgery probably outweigh the benefits. For patients with Barrett's esophagus, it is essential to continue the surveillance program even if the patient is asymptomatic after surgery, since progression of Barrett's esophagus after successful GERD surgery is well recognized.

Return to Meeting Publications