Stepping Up to the Complexities of Gastroesophageal Reflux

Between 4% and 7% of adult Americans report daily episodes of heartburn and/or regurgitation, approximately 20% experience these symptoms weekly, and half or more of adults encounter them at least once annually. Sixty percent of heartburn sufferers describe their discomfort as moderate to severe. Despite the prevalence and potential severity of these symptoms of gastroesophageal reflux, 35% of affected individuals do not seek medical care, preferring to self-medicate with non-prescription antacids or histamine receptor antagonists (H2RAs).

During this symposium, the faculty discussed the pathogenesis and diagnosis of gastroesophageal reflux and the current treatment options. Seymour M. Sabesin, MD, FACP (Rush Univer-sity, Chicago), moderator, characterized gastroesophageal reflux as a “chronic disease that is effectively and safely treated by inhibition of acid production by the stomach.”

This program was supported by an unrestricted educational grant from Eisai, Inc. and Janssen Pharmaceutica Inc.

The Physician Assistant’s Perspective on Gastrointestinal Reflux Disease

In many cases of gastroesophageal reflux, heartburn and regurgitation are accompanied by extraesophageal symptoms including chronic cough, sore throat, hoarseness, and may be associated with adult-onset asthma and other pulmonary disorders. These extraesophageal GERD-related symptoms may appear without significant complaints of heartburn or regurgitation. The severity of heartburn symptoms does not necessarily correlate with the presence or absence of visible esophageal inflammation or erosion, yet these symptoms may contribute to sleep disturbance and diminished health-related quality of life. Thus, in all cases, the primary objective of treatment for gastroesophageal reflux is symptom relief. Treatment of patients with erosive esophagitis, a condition for which the term “gastroesophageal reflux disease” (GERD) is often reserved, is also designed to prevent serious complications: esophageal strictures and ulcerations, and Barrett’s esophagus (ridges of metaplastic columnar epithelium in the distal esophagus) which is widely thought to be a precursor lesion for adenocarcinoma.

After describing the spectrum of symptoms with which patients with gastroesophageal reflux present (Figure 1), Kenneth D. Ingram, PA-C (Oregon Health and Science University) distinguished between patients with erosive esophagitis and those with nonerosive reflux disease (NERD). Until recently, most research on gastroesophageal reflux focused on erosive disease because investigators needed an “indisputable criterion for the diagnosis of GERD in clinical trials.” Endoscopic evidence met that need. However, Mr. Ingram said, this conceptual limitation “ignored accumulating evidence that the greatest impact of GERD is on the quality of life, and such impairment tends to be similar in patients with or without erosive esophagitis.” The assumption that therapeutic requirements are not as great in patients with NERD has also been contested in recent studies, emphasizing the clinical challenge and importance of this condition.

In light of this newer appreciation of the symptomatic and pathologic breadth of gastroesophageal reflux, “Physician assistants must recognize similarities and differences in clinical presentation between the erosive and nonerosive variants of GERD. They must also understand how best to evaluate patients with a variety of uncommon symptoms, all of which may be associated with this multifactorial disorder.”

Evaluating Gastroesophageal Reflux: Is it Important to Distinguish NERD from Erosive GERD?

Clifford S. Melnyk, MD, FACP, FACG, also from the Oregon Health and Science University, presented evidence suggesting that erosive GERD appears to be pathophysiologically distinct in endoscopic presentation and management from NERD. The pathophysiology of erosive GERD encompasses a series of major physiologic dysfunctions that may include impaired salivary clearance and mucosal defense, and failure of primary peristalsis resulting in impaired esophageal clearance. It also involves transient relaxations of the lower esophageal sphincter (LES) or the failure of an incompetent LES to prevent reflux of gastroduodenal contents. The consequences of poor esophageal function can be exacerbated further by delayed gastric emptying. Hiatal hernia may also contribute to GERD by depriving the esophagus of its “second sphincter.”

In contrast, patients with NERD present with lower incidences of primary peristalsis failure, LES incompetence, hiatal hernia, and esophageal acid exposure. Many of these patients experience bursts of postprandial regurgitation. Increased paracellular permeability or increased chemoreceptor sensitivity in esophageal epithelium may sometimes play decisive roles in the pathogenesis of NERD. It may be diagnostically important to remember that patients with NERD have a higher incidence of extra-esophageal symptoms, and may present without complaints of heartburn and regurgitation. A study using pH monitoring has shown that predominant symptoms of acid regurgitation and heartburn have a predictive value of 70% for erosive GERD, whereas inclusion of these in a broader complex of symptoms has a predictive value of 40% (regurgitation) or 46% (heartburn) for NERD (Klauser AF et al. Lancet 1990;335 (8683):205).

Although the ratio of NERD to erosive GERD is approximately 7:3, these variants of gastroesophageal reflux are not distinguishable by severity, chronicity, or duration of symptoms. Thus mild disease cannot be assumed to be NERD nor severe disease to be erosive GERD. A comparison of data from two studies in which heartburn grade (mild, moderate, severe) was correlated with patient-reported health-related quality-of-life parameters indicates, for example, that there is no significant difference between endoscopy-positive (N=550) and endoscopy-negative (N=845) populations (Venables TL et al. Scand J Gastroenterol 1997;32:965 and Carlson R et al. Eur J Gastroenterol Hepatol 1998;10:119). Consequently, the extent to which the symptoms of these two conditions overlap presents a diagnostic challenge.

One widely used diagnostic method for the presence of GERD as the basis for symptoms is a therapeutic trial of a proton pump inhibitor (PPI). Small trials using lansoprazole 30 mg once daily for 2 weeks, omeprazole 40 mg once daily for 2 weeks, and omeprazole 60 mg daily (20 mg in the morning and 20 mg at night) for 1 week have reported sensitivities to erosive GERD of 81%, 68%, and 83%, respectively, though specificity was not as high in any of the studies (Saxena A et al. Gut 1999;44:A112; Schenk BE et al. Am J Gastroenterol 1997;92:1997; Fass R et al. Gastroenterol 1997;113:A114). Thus if a patient responds to a PPI trial, no further diagnostic testing may be indicated at that time. Similarly, empiric trials of PPIs have shown a relatively high sensitivity for NERD (66% to 83%). Although the clinical response to a PPI trial does not distinguish between erosive GERD and NERD, the distinction is less important in responders than in nonresponders. Failure of the PPI trial requires additional investigation, quite likely to include 24-hour acid monitoring and
endoscopy with or without biopsy, to determine the presence or absence of esophageal erosion, strictures or ulcers, Barrett’s epithelium, or neoplasia.
PPI therapy is highly efficacious in erosive GERD. A meta-analysis of approximately 7,500 patients involved in multiple studies ranging in duration from 2 weeks to 12 weeks determined, for example, that by week 12, the healing rate associated with PPIs was 84% compared with 52% with H2RAs and 28% for placebo (Chiba N et al. Gastroenterol 1997;112:1799). PPI efficacy is also high in the management of NERD. Lind and colleagues have demonstrated, for example, that omeprazole 20 mg and omeprazole 10 mg induced a complete absence of heartburn in 46.3% and 31.1% of patients compared with 13.3% for placebo at week 4 (Lind T et al. Scand Gastroenterol 1997;32:974). Using the higher dose and removing the placebo effect yields a 33% complete-relief rate at 4 weeks that increases to 65% after week 12. Doubling the dose of a PPI does not dependably improve the response rate in symptomatic heartburn associated with NERD, however, because of the disease’s heterogeneous characteristics. In one subtype, pathological acid reflux occurs without gross esophagitis. In a second, hypersensitive esophageal mucosa continues to react to short-term bursts of postprandial reflux that may be technically within the normal range of esophageal acid exposure. And in a third subtype, heartburn is associated with functional symptoms such as nausea, fullness, and bloating that may or may not be resolved during PPI therapy.

An acid secretory study demonstrated that erosive esophagitis patients had gastric acid secretion of 56±8 mM (pH-1.25). In contrast, the mean postprandial gastric acid production in endoscopy-negative NERD patients was 110±16mM (pH=0.96) (Robinson M et al. Gastroenterol 2000;116:A2806). These different patterns may assist in distinguishing between the pathophysiologic processes in erosive GERD and NERD, and postprandial pH monitoring is essential for confirming reflux in those patients with symptoms who have grossly negative endoscopy and/or esophageal biopsies.

Alarm symptoms in gastroesophageal reflux patients—those that require a full diagnostic work-up—include dysphagia, involuntary weight loss, persistent vomiting, anemia, and a family history of upper gastrointestinal malignancy. In the absence of an endoscopically-measurable lesion to heal, symptom relief is the only outcome indicative of treatment success with NERD. Achieving symptomatic relief may be more difficult with NERD than with erosive GERD.

Controlling GERD Symptoms: The PPIs Are Not All Equal

Suppressing esophageal acid exposure is critical to achieving adequate symptom relief and enhancing patients’ well-being in both NERD and erosive GERD, and for promoting healing of esophageal inflammation and erosion in GERD. Clinical evidence indicates that PPIs surpass other drug classes in achieving these therapeutic objectives, although no agent in the PPI class can reliably achieve a 24-hour pH of 7.0. Control of pH and of symptoms is always variable with existing agents.

Malcolm Robinson, MD, FACP, FACG (University of Oklahoma and the Oklahoma Foundation for Digestive Research) compared the PPIs currently available in the United States using specific criteria for determining efficacy of treatment for controlling acute acid-mediated symptoms: prolonged elevation of pH to a level of 4.0 or above; reduction of acid volume; rapidity of onset of action and extent of day 1 symptom relief; duration of acid suppression; extent of inter-individual variation of efficacy; efficacy of a single dose level; absence of tolerance or tachyphylaxis; and overall safety with respect to drug-drug interactions and other adverse effects during long-term exposure.

All PPIs are prodrugs that undergo catabolic transformation to active sulfenamides, and the rapidity of this transformation governs the onset of action. Extensive comparative pharmacologic study indicates that this and other characteristics of proton pump inhibition differ among the PPIs. Kromer and colleagues have demonstrated that at a pH of 1.2 (that of an activated parietal cell), rabeprazole, omeprazole, lansoprazole, and pantoprazole all activate rather rapidly. At a pH of 5.1, however, a level typically reached when parietal cells are not secreting acid actively, there is wide variation in the rapidity of onset of drug action ranging from approximately 5 minutes for rabeprazole to as much as 282 minutes for pantoprazole (Kromer W et al. Pharmacol 1998;56:57). Omeprazole and lansoprazole require approximately 90 minutes for activation at the higher pH level. These varying activation rates correlate directly with speed of onset of effective acid control by these individual PPIs. In a study that measured proton pump inhibition at 5 minutes (rabeprazole) or 10 minutes (lansoprazole, omeprazole, and pantoprazole) and 45 minutes following administration, rabeprazole produced 100% pump inhibition at both time points compared with 66% and 100% for lansoprazole, 47% and 83% for omeprazole, and 20% and 49% for pantoprazole (Besancon M et al. Biol Chem 1997; 272:22438). The results of this study are shown in Figure 1.

There are also significant differences among the PPI agents with regard to overall first-day acid inhibition. In a study of 24-hour gastric pH, rabeprazole achieved a median intragastric pH of 3.4 (p=<0.05) compared with 2.9 for lansoprazole, 2.2 for pantoprazole, 1.9 for omeprazole, and 1.3 for placebo (Pantoflickova D et al. Gastroenterol 2000;118: A1290). In the same study, the median time during which the intragastric pH remained above 4.0 ranged from a high of 8.3 hours (p=<0.50) with rabeprazole to a low of 2.9 hours with omeprazole. With a median of 7.5 hours of pH control for the 24-hour period, only lansoprazole approached rabeprazole numerically. Nonetheless, no PPI achieves maximal acid inhibition on day 1. In a comparative study, Gardner and colleagues demonstrated that after a first dose, rabeprazole achieves 88% of maximal median inhibition compared with 42% for omeprazole (Gardner JD et al. Am J Gastroenterol 1999;94:2608). This inhibition is sustained well into the night with less acid breakthrough seen with rabeprazole than has been typical with other PPIs. Moreover, after 8 days of treatment, the effect of rabeprazole compared with omeprazole, as measured by the median time gastric pH remains above 4.0, remains significantly greater (p=<0.001) (Williams MP et al. Aliment Pharmacol Ther 1998;12:1079).

The foregoing effects of PPIs on intragastric pH significantly influence esophageal pH in reflux patients. In a single-agent study using two doses of rabeprazole, most subjects with abnormally high esophageal pH were normalized in terms of esophageal acid contact as of day 1, and all were normalized by day 7 of treatment (Robinson M et al. Aliment Pharmacol Ther 1997;11:973). In a placebo-controlled study comparing two dose schedules of rabeprazole with standard omeprazole therapy, both rabeprazole regimens were associated with superior normalization of esophageal pH (Galmiche JP et al. Aliment Pharmacol Ther 2001;15:343).

Importantly, these differences translate into marked differences in symptom relief over time as has been demonstrated by several studies. In one placebo-controlled study of omeprazole, the 4-week rate of relief from heartburn was 57% and from regurgitation was 75% compared with 19% and 47%, respectively, for placebo. Forty-three percent of treated patients were symptom free compared with 14% of patients taking placebo (Bate CM et al. Aliment Pharmacol Ther 1996;10:547). In a placebo-controlled study of two dose levels of rabeprazole, the median times to the first 24-hour heartburn-free period were 2.5 days and 3.5 days (placebo=19.5 days) and the complete heartburn resolution rates were significantly higher than for placebo. With both doses of rabeprazole, the time to the first heartburn-free night was 1.5 days compared with 7.5 days with placebo. All findings were statistically significant (Kahrilas PJ et al. Gastroenterol 2002;122:A1280). In another controlled study, one that evaluated a spectrum of symptoms, two dose levels of rabeprazole were associated with significant improvement in daytime and nocturnal heartburn, regurgitation, belching, bloating, satiety, and nausea (Johnson J et al. Gastroenterol 2002;122: A1294).

Dr. Sabesin presented the case of a 42-year-old woman with a multi-year history of moderately severe heartburn occurring several times each week and occasionally at night. Despite some relief from antacids and H2RAs, she has recently developed associated symptoms of regurgitation and bloating. She
was referred to a gastroenterologist who, on endoscopy, found no evidence of esophageal erosions.

Dr. Robinson expressed the judgment that endoscopy might not have been mandatory for such a patient, and almost certainly not before a trial course of PPI therapy. Even if she had not had a full response to therapy, the best time to do an endoscopic examination and to biopsy the esophagus is after a suitable course of PPI therapy has healed esophageal erosions and inflammation.

In a discussion of alternative treatments, Dr. Robinson noted that H2RAs are effective in many patients at the outset, but that their efficacy wanes for some individuals as a consequence of tachyphylaxis/tolerance. These medications benefit only a minority of patients who require constant and chronic therapy. In contrast, because PPIs inhibit the final common pathway for acid secretion, this class is not subject to the development of tolerance.

Dr. Melnyk emphasized that PPIs work effectively only after the proton pump has been activated, so the appropriate time to take PPIs are before breakfast or before dinner. Effects of PPIs taken just prior to sleep are diminished due to the absence of stimulated proton pumps.

On the subject of the duration of PPI therapy, Mr. Ingram suggested that fully responsive and stable patients without erosive disease may be candidates for dose reduction and possibly for substitution of an H2RA inhibitor. Dr. Melnyk cautioned, however, that abrupt termination of PPI therapy may result in gastric hypersecretion. This rebound is probably related to parietal cell hyperplasia that has been observed with PPIs. Thus if a patient is stepping down to an H2RA, the PPI should first be tapered. He added that H2RAs are frequently ineffective during the rebound hypersecretion phase that follows long-term PPI therapy, and that relatively few patients can be successfully “stepped down” to an H2RA.

Dr. Melnyk cited a general assumption that untreated NERD will eventually evolve into erosive GERD, but recent data suggest otherwise. In a recent study, approximately 1,500 patients with heartburn and regurgitation underwent endoscopy. The results indicated that patients with NERD could live into their 80s without developing esophageal erosions. Similarly, it seems likely that most individuals who do not manifest Barrett’s esophagus on initial endoscopy may never develop it. This, in turn, suggests that endoscopy should be a one-time event except in patients having Barrett’s esophagus on first endoscopy. In a forthcoming consensus paper, an expert panel maintains that only those patients with Barrett’s esophagus and evidence of dysplasia at screening will require continuous, long-term surveillance endoscopies.