Transient relaxation of the lower esophageal sphincter (LES) is an important cause of reflux events both in normals (J Clin Invest 1980;65:256) and in pts with esophagitis (Nejm 1982;307:1547). Newer evidence suggests that the frequency of these transient relaxations is nothigher in GERD pts than normal but that when relaxations occur they are more likely to be associated with reflux of acid in GERD pts (Am J Gastro 2001;96:2569). Transient relaxation is notthe mechanism of GERD in hiatus hernia (GE 2000;119:1439).

Some episodes of reflux are due to abrupt increases in intra-abdominal pressure that overcomes LES pressure (Nejm 1982;307:1547). This is probably a greater factor in pts with more severe grades of esophagitis than in those with mild disease (Gut 1995;36:505).

Hiatal hernia has been convincingly implicated as a contributor to reflux. The right diaphragmatic crus, which encircles the LES, augments LES pressure, particularly with sudden increases in abdominal pressure that occur with coughing and inspiration. Hiatal hernia interferes with this increased LES pressure, and the larger the hernia, the smaller the benefit of the crus in preventing reflux (Ann IM 1992;117:977).

About 30% of pts with heartburn severe enough to require self-medication with antacids have sensitivity to esophageal acid infusion or balloon distension despite normal endoscopies and pH probes. A lower threshold for esophageal pain may play a role in this subgroup of pts (Am J Gastroenterol 1999;94:628).

Delayed gastric emptying is an important mechanism for GERD in a small subgroup of pts (GE 1981;80:285).

In progressive systemic sclerosis (scleroderma), impaired esophageal acid clearance due to poor motility in the distal esophagus and a hypotensive esophagus are the primary defects (Dig Dis Sci 1992;37:833).

Emotional stress (caused by stressful tasks like playing video games and doing math problems!) increases the subjective severity of reflux sx but notthe actual number or duration of reflux events. The data suggest that pts who are anxious or subjected to stress may perceive minimal esophageal reflux as major symptomatic events.

Reflux in smokers is most often due to the combination of a decreased resting LES pressure and the abrupt increases in intra-abdominal pressure associated with coughing and deep inspiration (Gut 1990;31:4).

Obesity is associated with GERD sx, and even moderate weight gain is associated with worsened sx (Nejm 2006;354:2340).

H. pylori infection limited to the antrum might worsen reflux by increasing acidity, but more commonly (especially in the developing world), Hp gastritis infects the antrum and body of the stomach, and the resultant gastritis reduces acid production and reduces the incidence of GERD (Am J Gastro 2004;99:1222). There is a strong inverse relationship between Hp infection and severe forms of GERD such as Barrett’s esophagus (Am J Gastro 2004;99:1213).

Heartburn (a retrosternal burning sensation) is the most common sx, for which pts will use many synonyms, including indigestion and sour stomach.

Gross regurgitation of gastric contents, particularly after bending or with recumbency, is often incorrectly reported by pts as vomiting. It is distinguished from vomiting by the lack of the intense nausea that generally precedes vomiting, the lack of force, and the seemingly spontaneous nature of the sx. Minor episodes might only leave a bitter, hot taste in the mouth, especially upon awakening.

Dysphagia, the sensation that food sticks in the chest (p 6), is one of the most helpful clues to a dx of GERD if pts have primarily high epigastric pain or nausea.

Odynophagia, painful swallowing, is less common than dysphagia and suggests the possibility of infections (candida, herpes) or pill esophagitis, especially if sx are of sudden onset.

Belching is a frequent cause of reflux events in normal pts.

Nausea is notrare, and intractable nausea can be the primary presenting complaint (Ann Intern Med 1997;126:704).

Chest pain may occur with or without other typical reflux sx and may be difficult to distinguish from cardiac chest pain by hx alone (p 54).

Hoarseness, repetitive throat clearing, sensation of fullness in the back of the throat (globus sensation), and chronic cough are associated with GERD (Gastrointest Endosc 1995;43:225; Clin Gastroenterol Hepatol 2003;1:333).

Asthma is frequently found in association with GERD with up to half of pts in asthma clinics having symptomatic GERD and up to 80% of asthma pts having abnormal pH probes (Chest 1998;114:275).

Water brash, the abrupt, episodic filling of the mouth with large volumes of saliva, is an uncommon sx.

Technique for pH probe: (Am J Med 1997;103:130S; GE 1996;110:1982). In ambulatory pH probe studies, a pH probe is passed through the nostril and positioned 5 cm above the manometrically determined LES. The pH is recorded every 6-8 seconds. The accurate placement of the probe is crucial. The extra step of determining the location of the LES with manometry is necessary because determining the LES location by fluoroscopy, endoscopy, or seeing the step-up in pH that occurs when the catheter is pulled back from stomach to esophagus is notreliable (GE 1996;110:1982). Pts usually find the probe tolerable, though notpleasant. It is attached to a recorder worn by the pt. The pt indicates the occurrence of events such as meals, sleeping, or sx (heartburn, chest pain, cough) by pushing an event marker on the device and by means of a diary. Pts who pull the probe out after an hour because they cannot stand it are also providing their physicians with valuable information about their tolerance of discomfort. A wireless system using a pH telemetry device placed endoscopically (Bravo pH System) is an alternative to an indwelling catheter that is usually better tolerated (Am J Gastro 2003;98:740). Longer recording periods (48 hr) give the wireless system greater sensitivity (Clin Gastroenterol Hepatol 2005;3:329). Reflux episodes are defined as drops in pH to below a value of 4.0. Commercial probes provide information on percent total time pH <4, percent upright time pH <4, percent recumbent time pH <4, total number of episodes, and number of episodes longer than 5 min. The most important of these is percent time pH <4, which discriminates between abnormal and physiologic reflux. Most of the clinical application of pH probe is in the correlating of sx with reflux episodes.

Indications for pH probe: Only a tiny proportion of pts with reflux need pH probe as part of their evaluation. A pH probe is indicated for (1) proving abnormal reflux prior to sending a pt with a negative endoscopy to antireflux surgery, (2) evaluating recurrent reflux sx in pts who have had antireflux surgery, (3) evaluating sx that persist despite a month of a PPI, (4) evaluating noncardiac chest pain or throat sx that do notresolve with a PPI trial, and (5) evaluating whether reflux episodes are associated with adult-onset nonallergic asthma (GE 1996;110:1981).

Intraluminal impedance monitoring: This technique measures electrical impedance between pairs of electrodes attached to a plastic catheter in the esophageal lumen. During reflux of liquid, impedance falls rapidly; during reflux of gas, it rises rapidly. Thus impedance monitoring can provide information about reflux unrelated to acidity and may be of value in evaluating pts with sx unrelieved by PPIs or with atypical sx. Its clinical application has yet to be clearly defined (Am J Gastro 2007;102:187).

Omeprazole test: The response of sx to a 14-day trial of omeprazole 40 mg daily in pts with sx suggestive of GERD has a positive predictive value of 68% and a negative predictive value of 63% when compared to pH probe as a gold standard (Am J Gastroenterol 1997;92:1997). The fact that this number is nothigher reflects problems using pH probe as a gold standard. As a practical matter, if it sounds like GERD and gets better with omeprazole, it is probably GERD.

Indications for EGD: Pts with typical GERD sx who respond to a course of rx usually do notbenefit from endoscopy (Gastrointest Endosc 1999;49:834). EGD is clearly indicated in pts with dysphagia so that strictures can be found and dilated. EGD is needed in pts who show evidence of GI blood loss, iron deficiency, or sx suggestive of malignancy, such as weight loss. Many physicians offer EGD to pts with sx of several years’ duration because they are at a higher risk of developing Barrett’s esophagus. The benefit of this widely used approach is unproven (see Barrett’s esophagus p 41). Endoscopy is also appropriate for pts whose sx do notrespond to a course of rx. Endoscopy in this group may demonstrate esophagitis, stricture, or Barrett’s, or there may be an alternative explanation of sx, such as PUD, gastric outlet obstruction, gastroparesis with retained food, malignancy, candida, or herpetic esophagitis.

Grading of esophagitis: Several grading systems for esophagitis exist and serve as important tools for clinical studies. Grades are commonly used in endoscopy reports. The Savary-Miller system (Gastrointest Endosc Clin N Am 1994;4:677) and the Los Angeles classification (Gut 1999;45:172) are 2 common grading systems.

Endoscopic bx: Bx should be performed in pts with endoscopic esophagitis if they are immunosuppressed (Nejm 1994;331:656). Bx is less likely to be helpful if there is visible erosive change in a normal host. Erosions may obscure the presence of Barrett’s mucosa that may be detected by bx. Pts with grossly normal esophageal mucosa may benefit from bx. If bx shows convincing histologic evidence of reflux, the pt may be spared a pH probe for definitive dx. Intraesophageal eosinophils are perhaps the most useful diagnostic findings in reflux, though basal zone thickness and height of papillae are also helpful if specimens are properly oriented (Gastroenterol Clin North Am 1990;19:631).

Patient teaching: There is no simple therapeutic recipe appropriate for all pts with GERD. The most common error is failing to appreciate the severity and chronicity of a pt’s complaints and offering inadequate rx. Rx begins by teaching pts the basic pathophys of GERD so that they understand why their sx are likely to be chronic without appropriate rx. Most physicians suggest diet and lifestyle modifications despite the lack of critical evidence that such measures change outcomes (Am J Gastroenterol 2000;95:2692). Pts should be selectively advised to (1) avoid large meals, (2) avoid eating within 3 hr of bedtime, (3) stop smoking, (4) lose weight, and (5) elevate the head of the bed by 6 inches if nocturnal sx are problematic. Diet can be modified to minimize high-fat foods, chocolate, caffeine, citrus, alcohol, and tomato products (Mayo Clin Proc 2001;76:1002). Coffee produces reflux more than tea containing equal caffeine concentrations. Decaffeination reduces the
reflux caused by coffee, probably by removing substances in coffee other than just the caffeine (Aliment Pharmacol Ther 1994;8:283). The milder the GERD sx, the more likely that these modifications will make an impact on the need for drug therapy.

Drug rx: There are 2 approaches to drug rx. In some cases, it is best to start with a PPI for rapid sx relief and to switch to an H2RA after sx are completely relieved. Alternatively, one can begin with H2RA and intensify rx if sx are notrelieved. Severity of sx and cost of rx are major factors.

Promotility agents: Promotility agents are appealing in theory because they more directly address the pathophysiology of reflux. Delayed gastric emptying, inadequate LES pressure, and esophageal acid clearance are all potentially reversible with the ideal agent. As first-line rx, cisapride (Propulsid) was shown to be as effective as H2RAs for mild disease (Arch IM 1995;155:2165), and it was used in maintenance rx (Nejm 1995;333:1106). However, cisapride was withdrawn from the U.S. market because it caused life-threatening arrhythmias related to drug interactions and because of the drug’s effect in prolonging the QT interval. Metoclopramide (Reglan) should be avoided because of its limited effectiveness and the possibility of serious irreversible movement disorders (mostly tardive dyskinesia) (Arch IM 1989;149:2486).

Sucralfate: This agent binds to ulcer bases and binds to pepsin and bile salts. It has efficacy similar to H2RAs in some trials (Am J Med 1991;91:2A). Its most frequent use may be in pregnant pts where its minimal absorption is perceived as a possible benefit.

Maintenance therapy: GERD is a chronic disorder that usually requires chronic therapy. Maintenance rx is sensible (1) to control sx of relapsing disease or (2) to prevent complications of erosive esophagitis. For pts with symptomatic disease but minimal endoscopic esophagitis, intermittent rx with either omeprazole or ranitidine works well for about half of pts (BMJ 1999;318:502). Rapid recurrence of sx or slow response to retreatment should lead to continuous maintenance rx with the lowest dose of an antisecretory agent that controls sx.

PPI safety: PPIs are generally safe and well tolerated. Because of decreased gastric acidity, pts may be more susceptible to small intestinal bacterial overgrowth (Gut 1996;39:54) or acute infectious gastroenteritis. PPI therapy appears to be a risk for C. diff colitis (CMAJ 2004;171:33) and community-acquired pneumonia (Jama 2004;292:1955). Long-term PPI is associated with a greater risk of hip fracture (Jama 2006;296:2947) perhaps by decreasing calcium absorption. Because hip fracture is an infrequent event, the absolute hip fracture risk for an individual pt on long-term PPI therapy is an increase from 0.18% to 0.4% annually (GE 2007;132:2063). Ensuring adequate dietary calcium seems sensible.

Refractory GERD sx: Some pts will notrespond to conventional once-a-day dosing of PPIs with the relief of sx. In this case the dx should be carefully reconsidered. Malignancy, esophageal infection, gastroparesis, achalasia, and heart disease must be excluded. The most common reason for PPI failure is improper timing of doses.

Surgical therapy for GERD: While medical rx offers excellent relief of sx for the majority of pts, surgical rx is a consideration in several circumstances. Surgery should be considered in (1) pts whose sx are notadequately controlled by medical rx or (2) pts whose sx are well controlled by PPIs but who are intolerant of them or (3) pts who would rather have surgery than take medications chronically and who are willing to accept the greater risks of surgery.