Two experts in evidence-based medicine (G.T. and M.L.C.) undertook a systematic review (systematic search, critical appraisal and summary of the evidence) of treatment for GERD, based on an extensive, systematic review of the literature for the 3-year period from 1 January 2004 to 31 December 2006 to supplement the literature search conducted for the original Latin American publication. A systematic review and meta-analysis were performed but, unlike in the previous guideline, there was no formal voting to reach a consensus.

The overall review strategy was identical to that used for the original publication; a systematic, recursive literature search was carried out using Medline, the Cochrane Library and the Latin American and Caribbean Health Science Literature Database. The present guideline is intended for application in adults with GERD. For this reason, the search strategies aimed to retrieve studies performed in this age group; publications on Barrett's esophagus, extra-esophageal reflux disease, atypical GERD symptoms, and pediatric GERD were excluded. Studies addressing medical, endoscopic or surgical therapy were included if they assessed any of the following outcomes: heartburn relief, healing of esophagitis, heartburn remission, esophagitis relapse, and heartburn relapse; quality of life, satisfaction, and length of hospital stay were also included for studies of surgical treatment. The search was restricted to articles published in English, French or Spanish. The main search terms included gastroesophageal reflux (MeSH term), heartburn (MeSH term), and esophagitis (MeSH term); the types of publication included were meta-analyses, reviews, and randomized controlled trials.

The quality of the methodology was evaluated based on the users’ guide for medical literature for therapeutic and prevention studies.3 Levels of evidence and recommendation grades were defined according to the classification of the Oxford Centre for Evidence-Based Medicine4 (Table 1). Grade A is “highly recommended” and is applied to studies with grade 1 evidence (systematic reviews of randomized controlled trials or large randomized controlled trials with a low probability of bias or without bias).

The results of the intervention studies were synthesized in a meta-analysis with the Software Review Manager Version 4.2.5 and presented as relative risks (RR) and relative risk reductions (RRR), with their respective 95% confidence intervals (CI).

The data on treatment outcomes are presented in the following treatment categories: lifestyle and dietary interventions, short-term pharmacological therapy, long-term pharmacological therapy, endoscopic therapy, and surgical therapy. These categories are subdivided further, for pharmacological therapy, according to whether or not patients have confirmed erosive esophagitis – patients treated empirically, without prior endoscopy, or those with non-erosive reflux disease (NERD) and patients with erosive esophagitis – and to the type of therapy – antacids, alginates and sucralfate, histamine H2-receptor antagonists (H2-RA), prokinetics and proton pump inhibitors (PPI). In addition, on-demand therapy is discussed separately in the section on maintenance therapy, as the outcome measures are somewhat different and the study populations included patients with and without esophagitis. For each section, the update includes an overview of the background information, the outcome of the updated literature search, the implications for any recommendations and the current recommendation incorporating any necessary revisions.

The target users of the guideline will be gastroenterologists, family physicians, nurses, practitioners and all professionals involved in the care of GERD patients. The results will be presented in national meetings in Latin America and in the Panamerican Congress in 2010.

Potential barriers to implementation of the recommendations may include the costs of medications, insurance plans, the availability of diagnostic tools and drugs, and language and cultural barriers. Costs were not addressed.

The literature search, for the period from 1 January 2004 to 31 December 2006, identified 136 studies and 32 systematic reviews relevant to the treatment of GERD; of these, 79 met the specified inclusion criteria.

1. Diet and lifestyle changes

Overview: The major change in this area is increasing epidemiological evidence for an association between obesity and GERD and its complications.6 This change is of particular importance given the increasing prevalence of overweight and obesity in many parts of the world, including Latin America.7 A recent study, in randomly-selected women participating in the Nurses’ Health Study, indicates that weight gain is associated with an increased risk of GERD symptoms.8 However, there is limited, if any, data to confirm that lifestyle changes, dietary modification or weight loss improve GERD symptoms or healing of esophagitis,9 although there are some reports that the marked weight loss achieved by bariatric surgery may reduce reflux symptoms in some individuals.10

Review outcome: No new relevant studies on dietary or lifestyle modifications (e.g. raising the bed head during sleep) were published during the review period.11–25

Impact on recommendations: None.

Recommendation: The recommendation is unchanged with respect to the 2004 Consensus (see Table 2)

2. Pharmacological approach – short term therapy

a. No documented erosive esophagitis

i) Antacids, alginates and sucralfate

Overview: Antacids and antacid-alginate combinations are widely used, generally as over-the-counter preparations, for the management of GERD symptoms. Sucralfate is not widely used but, anecdotally, this drug may be prescribed in patients intolerant of or unresponsive to standard acid suppression therapy.

There are some data to indicate that antacid-alginate combinations are effective in short-term symptom control in patients with GERD, but therapeutic gain is small. There are insufficient data to support a recommendation that sucralfate be used for the treatment of GERD. Review Outcome: No relevant studies on antacid, alginates or sucralfate were published during the review period.

Impact on recommendations: None

Recommendation: The recommendation is unchanged with respect to the 2004 Consensus (see Table 2).

ii) Prokinetics

Overview: Medications with prokinetic activity, including metoclopramide, domperidone, cisapride, mosapride and tegaserod have been investigated as sole therapy and in combination with antisecretory medications for the treatment of GERD symptoms. Cisapride and tegaserod have been withdrawn from the market and mosapride is not widely available, leaving metoclopramide and domperidone as the only prokinetic agents available for general use.

Review outcome: No relevant studies were published on prokinetic agents during the review period.

Impact on recommendations: None.

Recommendation: The recommendation is unchanged with respect to the 2004 Consensus (see Table 2).

iii) H2-Receptor antagonists

Overview: H2-RAs are extensively used as over-the-counter and prescription therapy for the treatment of GERD symptoms. These agents are more effective than placebo for the relief of symptoms as empiric therapy and also for NERD; however, they are less effective than PPIs.1,26

Review outcome: No relevant studies on H2-RAs were published during the review period.

Impact on Recommendations: None.

Recommendation: The recommendation is unchanged with respect to the 2004 Consensus (see Table 2).

iv) Proton pump inhibitors

Overview: PPIs are generally accepted to be significantly more effective than placebo, H2-RAs and prokinetics for heartburn relief and general improvement of symptoms in patients without documented erosive esophagitis. Virtually all clinical data are derived from studies using a once-daily PPI and, although there are gastric pH data to show that twice-daily PPIs produce greater acid suppression than once-daily therapy in patients with GERD symptoms,27 there are no clinical data on high-dose PPI therapy in this patient group. Heartburn relief with standard dose omeprazole (20mg daily) was greater than with half-dose omeprazole (10mg daily) but previous studies have reported no differences between distinct doses of esomeprazole (40mg and 20mg) or between esomeprazole (40mg daily) and omeprazole (20mg daily).1

Review Outcome: Only one study comparing PPIs with placebo and utilizing one of the specified outcomes was published during the review period28; this study reported similar results to those of previous studies. Another study, comparing esomeprazole 40mg daily and esomeprazole 20mg daily with placebo29 in uninvestigated GERD patients, reported that esomeprazole reduced nocturnal heartburn, improved sleep quality and enhanced work productivity compared with placebo but that there was no difference between the two doses of esomeprazole.

Studies comparing PPIs with H2-RAs published during the review period30–33 confirm the superiority of PPIs for symptom relief in patients without documented erosive esophagitis (Fig. 1).

Fig. 1.

Short-term therapy in patients with NERD. PPI vs H2RA. Heartburn remission.

(0.61MB).

A study comparing PPIs in patients without documented erosive esophagitis34 reported no difference in symptom relief between esomeprazole (20mg daily) and pantoprazole (20mg daily) with RR values of 1.03 (95%CI: 0.89–1.20) and 1.01 (0.93–1.10) at 2 and 4 weeks, respectively.

Impact on recommendations: None.

Recommendation: PPIs are superior to H2-RAs or prokinetics for the initial management of patients with GERD symptoms but no documented evidence of erosive esophagitis (grade A recommendation). Despite evidence of dose-related differences in acid suppression for individual PPIs and of differences in acid suppression among PPIs, there are no clinical data to show that dose escalation or a change of PPI is associated with a greater effect on symptom relief.

b. Erosive esophagitis

i) Antacids, alginates and sucralfate

Overview: Antacids, antacid-alginate combinations and sucralfate are not generally prescribed or recommended for patients with documented erosive esophagitis, although these drugs are self-administered quite frequently as over-the-counter medications to produce short-term relief in patients taking prescription acid suppression therapy (Ref: Jones. R et al., 2001 APT35).

Review outcome: No relevant studies on antacids, alginates or sucralfate were published during the review period.

Impact on recommendations: None

Recommendation: The recommendation is unchanged with respect to the 2004 Consensus (see Table 2).

ii) Prokinetics

Overview: Medications with prokinetic activity, including metoclopramide, domperidone and cisapride, have been investigated as sole therapy and in combination with antisecretory medications for the treatment of erosive esophagitis. Cisapride and tegaserod have been withdrawn from the market, leaving metoclopramide and domperidone as the only prokinetics available for general use.

Review outcome: No relevant studies comparing prokinetic agents with placebo or other therapies were published during the review period.

Impact on recommendations: None.

Recommendation: The recommendation is unchanged with respect to the 2004 Consensus (see Table 2).

iii) H2-Receptor Antagonists

Overview: H2-RAs are more effective than placebo for the treatment of erosive esophagitis but are less effective than PPIs.

Review outcome: No relevant studies comparing H2-RAs with placebo or other therapies were published during the review period.

Impact on recommendations: None.

Recommendation: The recommendation is unchanged with respect to the 2004 Consensus (see Table 2).

iv) Proton pump inhibitors

Overview: PPIs are generally accepted to be significantly more effective than placebo, H2-RAs and prokinetics for the treatment of erosive esophagitis. These conclusions are based on extensive meta-analyses36 and are supported by numerous recent GERD consensus publications1,37,38. There is continuing controversy as to whether there are any differences among PPIs in terms of clinical outcomes in patients with erosive esophagitis. There is no evidence that the efficacy of PPIs is related to anything other than the degree and duration of acid suppression produced by these drugs. Previous studies have reported that esomeprazole (40mg daily) is more effective than omeprazole (20mg daily) and lansoprazole (30mg daily) in healing erosive esophagitis and relieving reflux symptoms in patients with erosive esophagitis, particularly in those with more severe esophagitis (Los Angeles [LA] grades C and D). Although the differences in outcome are likely to be related to the differing effects of the treatment regimens on acid suppression, the extent to which these differences in outcomes are medication-related or dose-related remains controversial and the absolute magnitude of any difference is difficult to estimate because of the confounding effect of other factors such as esophagitis severity.

Review outcome: The only study39 comparing a PPI with placebo in patients with erosive esophagitis used intravenous pantoprazole and did not evaluate any of the clinical outcomes specified in this analysis. No studies comparing PPIs with H2-RAs or prokinetics for the healing of erosive esophagitis were published during the review period.

However, several studies were published during the review period that compared distinct PPIs in the treatment of erosive esophagitis. A comparison of rabeprazole (20mg daily) and omeprazole (20mg daily) reported no significant differences in healing of esophagitis at 4 and 8 weeks 40; with respect to secondary endpoints, rabeprazole was superior in inducing remission of heartburn and lack of daytime pain.

A comparison of esomeprazole (40mg daily) and omeprazole (20mg daily) revealed no overall differences in healing of esophagitis at 4 weeks or 8 weeks,41 although esomeprazole produced healing in a greater proportion of patients with severe esophagitis (LA grade C and D) than did omeprazole. A meta-analysis of this new study with two prior randomized clinical trials42–44 comparing esomeprazole (40mg daily) with omeprazole (20mg daily) in 4818 patients with esophagitis showed that esomeprazole was more effective in producing healing of esophagitis at 4 and 8 weeks and in inducing remission of heartburn at 4 weeks (Fig. 2).

Fig. 2.

Short-term therapy in patients with esophagitis. Esomeprazole 40mg vs omeprazole 20mg. Healing of esophagitis at 4 weeks.

(0.32MB).

A randomized, controlled trial45 comparing lower dose esomeprazole (20mg daily) with omeprazole (20mg daily) reported no differences between the two PPIs, either in healing of esophagitis at 4 weeks (RR: 0.99; 95%CI: 0.92–1.07) and 8 weeks (1.03; 0.99–1.07) or in remission of heartburn at 4 weeks (1.00; 0.91–1.10).

Two randomized, controlled trials comparing esomeprazole (40mg daily) with pantoprazole (40mg daily) in patients with erosive esophagitis provided conflicting results; the first46 reported no differences between the two PPIs, whereas the second, much larger study47 reported that esomeprazole was significantly more effective than pantoprazole. A meta-analysis of these studies (Fig. 3a, b) indicates that esomeprazole was more effective than pantoprazole in healing of esophagitis at 4 weeks (RR: 1.08; 95%CI: 1.04–1.12) and 8 weeks (1.04; 1.02–1.06).

Fig. 3.

(a) Short-term therapy in patients with esophagitis. Esomeprazole 40mg vs pantoprazole 40mg. Healing of esophagitis at 4 weeks. (b) Short-term therapy in patients with esophagitis. Esomeprazole 40mg vs pantoprazole 40mg. Healing of esophagitis at 8 weeks.

(0.6MB).

A previous meta-analysis comparing esomeprazole (40mg daily) with other PPIs, including the data reported by Schmitt et al 41, concluded that esomeprazole produced a modest overall benefit in 8-week healing and symptom relief when all patients with erosive esophagitis were considered. The authors noted that the clinical benefit of esomeprazole appears to be negligible in less severe erosive disease but might be important in more severe disease48.

Impact on recommendations: None.

Recommendation: PPIs are more effective than placebo, prokinetics and H2-RAs in healing erosions and providing symptom relief in patients with erosive esophagitis. PPIs should be the initial therapy of choice (4–8 weeks) for erosive esophagitis. There appears to be little difference between PPIs in patients with mild esophagitis (LA grade A and B); however, healing may be produced in a higher proportion of patients with more severe erosive esophagitis (LA grade C and D) with esomeprazole (40mg daily) than with lansoprazole, omeprazole or pantoprazole if they have more severe erosive esophagitis (grade A recommendation).

3. Pharmacological Approach: Long-Term Therapy

Most long-term trials have evaluated the efficacy of continuous maintenance therapy, with daily drug administration, in the prevention of recurrent erosions, documented by endoscopy, or of recurrent reflux symptoms. However, in recent years, a number of studies have evaluated the efficacy of on-demand therapy for the management of recurrent reflux symptoms. This treatment paradigm requires that patients take medication only when they have recurrent symptoms and, as a result, evaluating treatment efficacy by documenting the prevention of recurrent symptoms is not appropriate. Because of concerns that non-continuous therapy would permit continuing esophageal injury, with the possibility of long-term adverse sequelae, most trials have been conducted in patients with NERD, complicating overall assessment of on-demand therapy. However, recent research has studied on-demand therapy in patients with erosive esophagitis, as well as those with NERD. As a result, it was decided that continuous maintenance therapy and on-demand therapy would be reviewed separately in this document.

a. No documented erosive esophagitis

i) Proton pump inhibitors

Overview: PPIs are generally more effective at preventing relapse of reflux symptoms than placebo, prokinetics or H2-RAs when prescribed as long-term therapy in patients without documented erosive esophagitis. However, the efficacy of long-term therapy depends, in part, on the treatment regimen: on-demand therapy allows patients to take medication as needed, when they have symptoms, intermittent therapy provides fixed-length courses of therapy (typically 2–8 weeks) in the event of symptom recurrence, while continuous maintenance therapy provides for uninterrupted, long-term daily medication intake.

Review outcome: One new study comparing on-demand lansoprazole (15mg daily) with placebo49 in patients with uninvestigated GERD reported that on-demand PPI therapy was superior to placebo with respect to control of heartburn and patient satisfaction. Although this study provided no direct data on heartburn recurrence, a Danish, multicenter, primary care study reported that on-demand therapy with esomeprazole (20mg daily, as needed) was associated with lower costs than intermittent, physician-controlled 2-week or 4-week courses of esomeprazole (40mg daily); on-demand therapy was associated with fewer episodes of relapse than intermittent therapy during the 6-month study but there were no major differences in patient satisfaction.50 However, when on-demand therapy (esomeprazole 20mg daily, as needed) was compared with continuous PPI therapy (esomeprazole 20mg daily), heartburn remission and quality of life were similar for the two PPI regimens, although both were more effective than continuous ranitidine therapy (150mg twice daily).51

Impact on Recommendations: The updated review is consistent with prior data that on-demand PPI therapy is superior to placebo and H2-RAs for long-term therapy in patients without documented erosive esophagitis. Patient outcomes are similar for on-demand, intermittent and continuous long-term therapy, although on-demand therapy may be associated with lower costs.

Recommendation: There is good evidence (type 1) supporting the use of PPIs instead of H2-RAs or prokinetics in the maintenance treatment of patients with GERD without documented erosive esophagitis. Consequently, patients needing ongoing treatment should be offered any PPI as a first choice maintenance therapy (grade A recommendation). The choice of treatment strategy is a matter for discussion between the patient and the physician, after consideration of costs, convenience and treatment aims. There are no data to indicate that one PPI is preferable to another in this patient group (grade D recommendation).

b. Erosive esophagitis

i) Histamine H2-Receptor antagonists

Overview: H2-RAs are widely used in the long-term management of GERD but, although the previous consensus publication noted that famotidine was more effective than placebo in maintaining remission from esophagitis at 6 months (RR: 0.18, 95%CI: 0.09–0.37), tachyphylaxis occurs with H2-RA therapy and consequently these drugs are significantly less effective than PPIs in preventing relapse of erosive esophagitis.

Review outcome: No relevant studies on long-term H2-RA treatment for erosive esophagitis were published during the review period.

Impact on recommendations: None.

Recommendation: The recommendation is unchanged with respect to the 2004 Consensus (see Table 2).

ii) Proton pump inhibitors

Overview: PPIs are the most effective long-term medical therapy for preventing recurrent erosions and reflux symptoms in patients with erosive esophagitis. The 2004 Consensus noted that, although H2-RAs (such as famotidine) are more effective than placebo (RR 0.18, 95% CI 0.09 to 0.37) in maintaining the healing of esophagitis at 6 months, PPIs are more effective than placebo, H2-RAs and prokinetics.

At the 2004 Consensus, it was noted that studies had shown no differences between rabeprazole 10mg and rabeprazole 20mg, rabeprazole 10mg, rabeprazole 20mg and omeprazole 20mg, lansoprazole 30mg and pantoprazole 40mg, lansoprazole 30mg and omeprazole 20mg, pantoprazole 40mg and omeprazole 20mg, pantoprazole 20mg and pantoprazole 40mg or esomeprazole 40mg and esomeprazole 20mg in the maintenance of endoscopic and clinical remission of patients with esophagitis. However, the finding that omeprazole 20mg was superior to omeprazole 10mg in maintaining healing of esophagitis does support the concept that the efficacy of maintenance therapy in this class of medication is related to the degree of acid suppression produced.

Review outcome: No relevant studies comparing PPIs with placebo for maintenance therapy in erosive esophagitis were published during the review period. However, a number of studies compared distinct PPIs. An updated meta-analysis that added the study by Devault52 to that of Lauritsen53 shows that esomeprazole 20mg is more effective than lansoprazole 15mg in maintaining endoscopic and clinical remission at 6 months (Fig. 4). Similarly, an updated meta-analysis that added the study by Caos54 to that by Thjodleifsson55 shows that rabeprazole 20mg daily is more effective than rabeprazole 10mg daily in preventing relapse of heartburn and erosions at 5 years (Fig. 5).

Fig. 4.

Maintenance therapy in patients with esophagitis. Esomeprazole 20mg vs lansoprazole 15mg. Remission at 6 months.

(0.4MB).

Fig. 5.

Maintenance therapy in patients with esophagitis. Rabeprazole 20mg vs rabeprazole 10mg. Relapse of esophagitis.

(0.39MB).

The EXPO study56 reported that esomeprazole is more effective than pantoprazole both in healing esophagitis after 4–8 weeks of treatment and in subsequently keeping patients in endoscopic and symptomatic remission at 6 months of maintenance therapy.

Impact on Recommendations: The new data do not affect the prior recommendation that continuous maintenance therapy with PPIs is superior to placebo, antacids and alginates, H2-RAs and prokinetics in preventing recurrent esophagitis and reflux symptoms. However, the updated meta-analyses do provide further support for the notion that the efficacy of PPI maintenance therapy is related to the degree of acid suppression produced by these drugs and that there is a dose-response relationship for PPIs in the long-term treatment of GERD.

Recommendation: There is good evidence supporting the use of PPIs instead of H2-RAs or prokinetics in the maintenance therapy of GERD patients (with or without erosive esophagitis) (grade A recommendation). Consequently, patients requiring continuous GERD therapy should be offered a PPI as first-line maintenance treatment. However, the data indicate that treatment outcomes are related to the degree of acid suppression achieved by PPI therapy; healing, and to a lesser extent, symptom resolution, are dependent upon the dose and frequency of PPI administration, as well as the choice of PPI. At lower doses, approved for maintenance therapy, esomeprazole is associated with lower relapse rates than lansoprazole or pantoprazole, especially in patients with more severe, LA grade C and ‘D’ esophagitis (grade A recommendation). For esomeprazole, as for other PPIs, dosage is also important; in this context, for example, rabeprazole 20mg daily is more effective than rabeprazole 10mg daily and esomeprazole 40mg daily is more effective than esomeprazole 10mg daily, although there is little difference between esomeprazole at 40mg and 20mg doses (grade A recommendation). Overall, full-dose PPI therapy is more effective than half-dose PPI therapy, although the clinical significance of these differences and of the differences between distinct PPIs remains unproven.

i) Proton pump inhibitors

Overview: In patients with NERD, on-demand PPIs are more effective than placebo in controlling reflux symptoms such that patients are satisfied with their treatment.57,58 There is evidence of a dose-response effect for PPIs, at least at lower doses. However, although omeprazole 20mg was more effective than omeprazole 10mg as on-demand therapy, there was no significant difference in subsequent trials between esomeprazole 40mg and esomeprazole 20mg.59

Review outcome: An important study by Sjostedt60 comparing daily versus on-demand therapy with esomeprazole 20mg in 470 patients showed no differences between treatment strategies in symptom control or satisfaction with therapy but endoscopic remission at 6 months favored daily administration.

Impact on recommendations: There were no specific recommendations in the previous consensus publication, although the document recognized that, in NERD patients, on-demand PPI therapy was more effective than placebo in symptom control. Subsequent data support this conclusion with the caveat that on-demand therapy is less effective than continuous, daily therapy in preventing relapse of erosive esophagitis.

Recommendation: On-demand PPI therapy provides effective symptom control in patients with NERD and mild erosive esophagitis but is not recommended for the prevention of recurrent esophagitis.

The measures below are recommended based on the 2004 Consensus, which has been modified, when appropriate, by new evidence that became available in the subsequent review period.