A Randomized Trial of Medical and Surgical Treatments for Patients With GERD Symptoms That Are Refractory to Proton Pump Inhibitors

Background: Gastroesophageal reflux disease (GERD), which affects at least 20% of adult Americans, may be especially common and severe in Veteran patients. Proton pump inhibitors (PPIs), which block gastric acid production, are the most effective medications for GERD, and the VA spends more than $177 million each year on outpatient PPI prescriptions. PPIs do not prevent the reflux of non-acidic material and do not completely eliminate esophageal acid exposure, however, and bothersome GERD symptoms persist in approximately 40% of patients treated with PPIs. Recent studies using the new technique of esophageal pH/ impedance monitoring, which detects the reflux of both acidic and non-acidic materials, have shown that PPI-resistant GERD symptoms correlate with episodes of reflux (acidic and/or non-acidic) in approximately one-half of patients. For those patients, an antireflux operation might relieve symptoms and obviate the expense of ineffective PPI therapy, but the efficacy of modern, laparoscopic fundoplication in this regard is not clear. For patients with PPI-resistant GERD symptoms, furthermore, the efficacy of medications that that can prevent gastroesophageal reflux (e.g. baclofen) or diminish pain of esophageal origin (e.g. neurotropic agents like desipramine) also is not clear. Study Hypothesis: Laparoscopic antireflux surgery (Nissen fundoplication) is superior to medical therapy (PPIs plus baclofen and desipramine) for GERD patients who, while on PPIs, have persistent episodes of heartburn that are associated with reflux episodes or with abnormal esophageal acid exposure by esophageal pH/impedance monitoring.

Study Goals: The primary goal is to compare the efficacy of laparoscopic Nissen fundoplication and medical therapy (PPIs plus baclofen and desipramine) for GERD patients who, while on PPIs, have persistent episodes of heartburn that are associated with reflux episodes or with abnormal esophageal acid exposure by esophageal pH/impedance monitoring, and to compare the efficacy of each therapy with placebo. Secondary goals are: 1) To determine the frequency with which non-GERD disorders underlie "PPI failure," 2) To determine the frequency of functional gastrointestinal symptoms, anxiety and depression in patients who have persistent heartburn while on PPIs, 3) To determine whether functional gastrointestinal symptoms, anxiety and depression is associated with the outcomes of medical and surgical therapies, and 4) To determine whether the outcome of Nissen fundoplication is associated with adherence to technical aspects of the operation.

Study Design: Up to 16 VA medical centers, there will be a 30-month recruitment period to enroll 108 patients with heartburn that is refractory to PPI therapy. Patients will have their baseline GERD symptoms scored using the GERD Health-Related Quality of Life (GERD-HRQL) index, and will have endoscopy, esophageal manometry and esophageal pH/impedance monitoring while on PPI therapy. Patients who have episodes of heartburn that are associated with reflux episodes or with abnormal esophageal acid exposure by esophageal pH/impedance monitoring will be randomized to one of three treatment groups: Surgical Treatment (laparoscopic Nissen fundoplication), Active Medical Treatment (omeprazole and baclofen initially; desipramine for baclofen failures) or Placebo Medical Treatment (omeprazole, placebo baclofen, placebo desipramine). All patients will have quarterly clinic visits for symptom scoring and laboratory testing. At one year, patients will have a final symptom scoring and repeat endoscopy, esophageal manometry and esophageal pH/impedance monitoring. Treatment success will be defined as 50% improvement in the GERD-HRQL score at 12 months. Patients also will complete the Hospital Anxiety and Depression Scale (HADS), Rome III Functional GI Disorders Questionnaire and the Short-Form Health Survey (SF-36) at baseline and one year. The results will be correlated with treatment outcomes.