cleveland clinic journal of medicine-2003-maton-s51

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 70 • SUPPLEMENT 5 NOVEMBER 2003 S51

■ ABSTRACT

The choice of a medical therapy to treat gastroesoph-ageal reflux disease (GERD) centers around severalfactors, including the efficacy and safety of theagent and the severity of the patient’s symptomsand complications. Although the efficacy of antacidsand alginic acid has not been proven definitively inclinical trials, these agents are effective against mildGERD symptoms in clinical practice. Along withsucralfate, these agents are also useful in specialpopulations, such as pregnant women, for whomacid-suppressive therapy may not be the bestoption. The withdrawal of cisapride from the USmarket has lessened the role of promotility agentsfor treating GERD, as their efficacy must beweighed against their side effects. Acid-suppressiveagents have become the drugs of choice for GERD.Both proton pump inhibitors (PPIs) and histamineH2-receptor antagonists effectively and safely treatGERD. However, PPIs have been shown to providethe highest levels of GERD symptom relief andesophageal healing to the most patients, in theshortest time, and with the fewest side effects.

Pharmacotherapy is considered first-linetreatment for patients with gastroesophagealreflux disease (GERD). Although someguidelines recommend instituting lifestyle

changes at the same time as an initial trial of empir-ic medical therapy,1 others note that these diet andlifestyle changes have little therapeutic benefit, andrecommend medical therapy as initial treatment.2

The following medical therapies are available for

the treatment of GERD:• Prokinetic agents, which target the underlying

motility dysfunction that causes GERD• Mucosal-protective agents—ie, sucralfate, which

binds with damaged mucosa to form a barrieragainst harmful acid reflux, and alginic acid,which forms a foamy barrier on top of the reflux-ate to protect the esophagus

• Acid neutralizers (antacids), which work locallyto raise the pH of the refluxate

• Acid-suppressive agents—ie, histamine H2-receptor antagonists (H2RAs) and proton pumpinhibitors (PPIs), which inhibit acid productionin the parietal cell.Guidelines also differ on which medical therapy

should be used as initial GERD treatment. Therapyfor any disease must be effective and safe, and fit theneeds of the patient. Other issues, such as concomi-tant conditions, recurrence of symptoms, and cost totreat, should also be considered when deciding on acourse of therapy.

This article examines the safety and efficacy ofthe available medical therapies for GERD. Endpoints for the efficacy of these agents include:• Symptom relief, which is a measure of the reduc-

tion in symptoms (usually heartburn and regurgita-tion, but some studies assess noncardiac chest painand other atypical or extraesophageal symptoms)

• Symptom resolution, which indicates the absenceof symptoms

• Erosive esophagitis healing rates. Maintenance studies have examined maintenanceof erosive esophagitis healing and symptom recur-rence. Safety considerations include adverse eventsand the effects of long-term treatment.

This article also presents data from placebo-con-trolled and comparative trials of the availableH2RAs and the available PPIs, as well as trials com-paring efficacy between H2RAs and PPIs. Much ofthe literature on GERD pharmacotherapy focuses onthe safety and comparative efficacy of these acid-sup-

Profile and assessment of GERD pharmacotherapy

PAUL N. MATON, MD

From the Digestive Diseases Research Institute, OklahomaCity, Okla.

Address: Paul N. Maton, MD, 3435 NW 56th Street, Suite206, Oklahoma City, OK 73112.

S52 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 70 • SUPPLEMENT 5 NOVEMBER 2003

G E R D P H A R M A C O T H E R A P Y

pressive agents. PPIs have been recognized as themost effective medical therapy for GERD symptomrelief, for healing all grades of erosive esophagitis,and for maintenance of healing. Although somepatients experience symptom relief and healing oferosive esophagitis with H2RAs, PPIs produce morefrequent and rapid symptom relief and esophagealhealing for a greater percentage of patients.1

■ PROKINETIC AGENTS

Prokinetic agents that treat GERD increase loweresophageal sphincter pressure (LESP), accelerategastric clearance, stimulate esophageal peristalsis,increase the amplitude of esophageal contractions,or perform a combination of two of these actions.

All prokinetic agents (bethanechol, metoclopra-mide, domperidone, and cisapride) are effective, tovarying degrees, in improving GERD symptoms andhealing esophagitis. However, efficacy data for theseagents come from small, sometimes poorly designedstudies, often without a placebo control. Also, theadverse-event profile of these agents must beweighed against any clinical benefit of GERD treat-ment. Although domperidone (available in Canadabut not in the United States) is well tolerated,metoclopramide and bethanechol have been associ-ated with significant adverse events (Table 1).3 Cis-apride, in particular, although the most effective ofthe prokinetic agents for treating GERD, wasremoved from the US market because of deathsassociated with cardiac arrhythmia.

BethanecholBethanechol is a direct-acting muscarinic receptoragent that acts by stimulating the parasympatheticnervous system to release acetylcholine. It has beenshown to increase LESP and improve esophagealperistaltic clearing.

Clinical efficacy. Some small, double-blind,placebo-controlled studies have investigated the effi-cacy of bethanechol in GERD treatment, with mixedresults. One placebo-controlled study conducted in20 patients found that a 2-month course of bethane-chol 25 mg four times daily reduced heartburn andreduced antacid use.3 However, in another study of44 patients by Thanik and colleagues,4 the improve-ment of GERD symptoms in patients receivingbethanechol plus antacids was not statistically signif-icantly different from that in patients receivingantacids plus placebo.

Results also differ among studies examining the

efficacy of bethanechol in healing erosive esophagitis.In a comparative trial of bethanechol and cimetidine,the two agents had fairly similar healing rates (52%of patients receiving bethanechol and 68% of thosereceiving cimetidine experienced complete heal-ing). Both agents were administered with high dosesof antacids, which may have helped produce thesehigh healing rates.3 Interestingly, although Thanikand colleagues4 found bethanechol to be no moreeffective than placebo in improving GERD symp-toms, 45.5% of patients receiving bethanechol 25mg four times daily experienced complete healing oferosive esophagitis, compared with 13.6% ofpatients receiving placebo plus antacids (P < 0.015).

Safety. Unfortunately, at the dosage level neces-sary to treat GERD (25 mg four times daily), beth-anechol can cause significant side effects, such asabdominal cramping, blurred vision, fatigue, andincreased urinary frequency. Side effects occur inabout 10% to 15% of patients, and are more com-mon in the elderly. Bethanechol is also associatedwith a long list of contraindications (Table 1) thatcompromise its use as an anti-GERD agent.3

MetoclopramideMetoclopramide is a dopamine antagonist. Al-though its precise mechanism of action is unclear, itseems to sensitize tissues to the action of acetyl-choline. It has been shown to increase the ampli-tude of gastric and esophageal contractions,increase LESP, and increase the speed of gastricemptying and intestinal transit.

Clinical efficacy. In two small, placebo-con-trolled studies in which 31 and 15 patients withGERD received metoclopramide 10 mg three timesdaily, symptom improvement did not differ signifi-cantly between the treatment and control groups.However, in studies conducted in 30 and 31patients with GERD, a higher dosage of the agent,10 mg four times daily, either alone or in combina-tion with an antacid, was more effective than place-bo at improving symptoms.5,6

Comparative studies have found that metoclo-pramide is as effective as H2RAs (cimetidine andranitidine) in relieving heartburn and other GERDsymptoms.7,8 All of these comparative trials wereconducted in small patient populations,3 and all butone were conducted without a placebo control.8

The largest one, conducted in 73 patients, found nodifference in symptom relief between patients givencimetidine 400 mg four times daily alone and those


M AT O N

given a combination of cimetidine with metoclo-pramide 10 mg three times daily.9

Although symptom improvement has beendemonstrated with metoclopramide, this agent doesnot seem to be significantly more effective than pla-cebo at promoting healing of erosive esophagitis.3 Inthe one placebo-controlled study comparing it withcimetidine, metoclopramide improved the appear-ance of esophageal erosions in 82% of patients, butthis was not significantly different from rate witheither cimetidine or placebo (78% for each).8 Inanother comparative study, both metoclopramideand ranitidine produced significant healing, butmetoclopramide was effective in fewer patients(52% healing rate, vs 81% with ranitidine).10 Therecommended dosage of metoclopramide is 10 mgfour times daily, whereas the recommended dosage ofranitidine is 75 mg twice daily.

Safety. To an even greater extent than withbethanechol, side effects are a significant drawbackto GERD therapy with metoclopramide. Because itis a centrally acting dopamine antagonist that cross-es the blood-brain barrier, antidopaminergic sideeffects are common, occurring in 20% to 30% ofpatients. Drowsiness and lassitude are most com-mon, and anxiety, agitation, confusion, hallucina-

tions, and motor restlessness have also been report-ed.3,11 The most serious effects are depression and tar-dive dyskinesia, which may be irreversible. Adverseevents are most common at higher doses and in chil-dren, young adults, and the elderly. Other less com-mon adverse events are listed in Table 1.

DomperidoneDomperidone is another dopamine antagonist, al-though it is not available in the United States. Itstimulates esophageal peristalsis, increases LESP,and accelerates gastric emptying.

Clinical efficacy. As with bethanechol andmetoclopramide, data on the efficacy of domperi-done in GERD treatment come from small studies.The largest one, conducted in 45 patients, com-pared domperidone and ranitidine without a place-bo control.

The efficacy of domperidone in GERD treatmenthas not been persuasively proven in well-controlleddouble-blind studies, and results with domperidoneat dosages of 20 mg three or four times daily areinconsistent.3 In one study, domperidone was nomore effective than placebo in reducing the numberof reflux episodes or improving GERD symptoms,although antacids were used less frequently at the

TABLE 1Contraindications and adverse events associated with prokinetic agents3

Agent Contraindications Adverse events

Bethanechol • Intravenous or intramuscular use may cause • Abdominal cramping, blurred vision, fatigue, severe cholinergic reaction and increased urinary frequency in 10% to 15%

• Hyperthyroidism, peptic ulcer, bronchial asthma, of patientsmechanical obstruction of gastrointestinal or lower urinary tract, peritonitis, parkinsonism, bradycardia, atrioventricular conduction defects, hypotension, and coronary artery disease

Metoclopramide — • Antidopaminergic side effects in up to 30% ofpatients, including drowsiness, lassitude, anxiety,agitation, and motor restlessness

• Dystonic reactions in 1% of patients• Parkinson symptoms (tremor, rigidity, akinesia, tardive

dyskinesia) rare except with high doses (30–80 mg)• Gynecomastia, galactorrhea, and menstrual disorders

Domperidone — • Hyperprolactinemia, resulting in breast enlargement,nipple tenderness, galactorrhea, and amenorrhea, in10% to 15% of patients

• Otherwise well tolerated

Cisapride — • Deaths associated with cardiac arrhythmia led tovoluntary removal from US market



end of the trial compared with baseline in the dom-peridone group.12 Other studies have shown dom-peridone to be effective in relieving symptoms butnot in healing esophagitis.3 In two non–placebo-controlled comparative trials of domperidone andH2RAs (ranitidine or famotidine), the two agentsproved to be similarly effective in symptom reliefand in promotion of esophageal healing. However,the combination of domperidone with an H2RAwas not significantly better than each agent givenalone.13,14

Safety. Although domperidone is a dopamineantagonist, it does not cross the blood-brain barrier(unlike metoclopramide) and was developed to actas a specific antagonist to the inhibitory effects ofdopamine on the gastrointestinal tract. It is well tol-erated, with few significant side effects. The adverseevents that do occur are related to the stimulationof prolactin release (Table 1) and are seen inapproximately 10% to 15% of patients. Theseevents can be seen with metoclopramide use but aremore common with domperidone because it isadministered in higher doses. Domperidone rarelycauses extrapyramidal side effects.3

CisaprideAny discussion of cisapride must be prefaced with anote on its profile and current market availability.High blood concentrations of cisapride can causeQT prolongation and cardiac arrhythmia, includingventricular arrhythmia, such as torsades de pointes.Coadministration of a number of drugs can reducehepatic metabolism of cisapride and increase thelikelihood of toxic concentrations.

Cisapride was removed from the US market inJuly 2000 after 341 cases of arrhythmia and 80deaths were spontaneously reported to the FDAfrom July 1993 to May 1998. The agent is nowavailable only on a restricted basis through a limit-ed-access program for patients who have failed torespond to or cannot receive alternate therapies.15

Cisapride acts locally on the gastrointestinaltract and seems to facilitate release of acetylcholinefrom postganglionic neurons in the myentericplexus. There is also evidence that it influences theactivity of other chemical mediators of mucosal andmuscular function in the gut, interacting with theserotonin 5-HT4 receptor in the myenteric plexus.Cisapride increases smooth muscle contractility,increases LESP, and enhances esophageal peristalticfunction.

Clinical efficacy. Before its removal from the USmarket, cisapride was indicated for supplementaltreatment of nocturnal heartburn symptoms. It wasthe most effective promotility agent available forthe treatment of GERD, in terms of both higherefficacy and fewer reported side effects. In clinicaltrials, cisapride was consistently better than placeboat improving the symptoms of GERD and promot-ing healing of erosive esophagitis. Optimal efficacyfor relieving symptoms was achieved at a dosage of10 mg three times daily, whereas 10 mg four timesdaily showed efficacy in healing esophagitis. Onestudy found that 10 mg of cisapride given four timesdaily was as effective as 20 mg given four times dailyin healing esophagitis.16

Comparative trials of cisapride and H2RAs yield-ed similar efficacy rates in healing esophagitis.3 Gal-miche and colleagues17 found that cimetidine 400mg and cisapride 10 mg, each given four times daily,produced endoscopic healing rates of 57% and 56%,respectively, in patients with erosive esophagitisgrades I to III (ie, mild to moderate esophagitis).(Here and except where noted otherwise, referencesto erosive esophagitis grades in this article are to theSavary-Miller classification system.)

■ MUCOSAL-PROTECTIVE AGENTS

SucralfateSucralfate is a mucosal-protective agent that binds toinflamed tissue, creating a protective barrier. It blocksdiffusion of gastric acid and pepsin across the barrierand inhibits the erosive action of pepsin and bile.18

Sucralfate is available in the United States in

TABLE 2Results of comparative trials of sucralfate in the healing of erosive esophagitis

Patients with healederosive esophagitis

Investigators N Weeks Sucralfate Comparator

Simon 41 8 64% 68%et al20 (ranitidine)

Hameeteman 42 8 31% 14%et al21 (cimetidine)

Laitinen 68 6 53% 34%et al22 (alginate +

antacid)


M AT O N

tablets and in suspension form. As suspension,sucralfate is administered in 1-g doses four timesdaily. Physicians rarely prescribe this agent to treatGERD, but it can be useful as an anti-GERD thera-py in special populations, such as in women who arepregnant. Although sucralfate contains aluminum,which can be harmful to a fetus, little systemicabsorption of the agent occurs. As a result, sucral-fate is considered safe enough for the treatment ofheartburn in pregnant women.19

Clinical efficacy. The findings of three compara-tive, non–placebo-controlled studies20–22 examiningthe effects of sucralfate in patients with all grades oferosive esophagitis are summarized in Table 2. Thetrials compared sucralfate with H2RAs or withalginic acid plus antacids. Patients in all three stud-ies had endoscopically confirmed reflux esophagitis,and the results of therapy were endoscopically con-firmed as well. The degree of healing with sucralfatecorrelated with the degree of injury, with highergrades of erosive esophagitis responding less favor-ably to treatment. In all three trials, symptomimprovement was rated as equally good for patientsin all groups. Esophagitis healing rates among thesucralfate and H2RA recipients were not statistical-ly significantly different in one trial20 but did differin another study,21 with more patients showinghealed or improved esophagitis in the sucralfategroup. In the remaining study,22 sucralfate generatedcomplete healing more often than did alginicacid/antacid. However, these studies are significant-ly limited by their relatively small size (40 to 70patients) and lack of a placebo control.

Other studies have been conducted in more spe-cific patient populations and have included a place-bo arm as a comparison. Chiba and colleagues23

pooled data from several studies of the erosiveesophagitis healing rates achieved with variousagents, including sucralfate, in patients with grades IIthrough IV esophagitis (Figure 1). Healing occurredin an average of 39% of patients who received sucral-fate, but this healing rate was accompanied by a verylarge 95% confidence interval (3.6% to 74.8%),indicating that it was not statistically significantlydifferent from the rate with placebo.

Sucralfate has also been studied in patients withGERD without erosive esophagitis. Simon and col-leagues24 tested the effects of sucralfate gel and pla-cebo in 141 patients with moderate to severe GERDbut with no esophageal erosions or ulcers. The over-all response rate after 6 weeks of treatment was 71%

with sucralfate, compared with 29% with placebo(P < 0.0001). Improvement in the maximum sever-ity of daytime and nighttime heartburn occurred in77% and 67%, respectively, of patients given sucral-fate, compared with 48% and 51%, respectively, ofpatients given placebo.

Alginic acidAlginic acid is often given in combination with anantacid. The first component provides a floatingbarrier on the gastric pool to minimize contactbetween gastric contents and esophageal mucosa,while the antacid temporarily neutralizes stomachacid.25 Like antacids alone, this combination thera-py helps to control mild to moderate reflux symp-toms in clinical practice.26 Tytgat and Nio11 notedthat improvement in GERD symptoms occurred inthree of four studies that compared alginate/antacidcombination therapy with placebo. However, whencompared with antacids alone, the alginate combi-nation therapy was superior in only one of four stud-ies. Convincing proof of esophageal healing hasnever been obtained in any study, and alginic acidtherapy is probably no better than antacid therapyin treating moderate to severe GERD.26

■ ANTACIDS

Antacids are the most widely used agents for treatingGERD because patients with mild heartburn oftenself-medicate with these over-the-counter drugs andnever seek treatment for their reflux symptoms.Available in liquid and tablet forms, antacids areused as needed. Some patients use antacids to sup-plement other anti-GERD therapies. In clinical

*Cimetidine, ranitidine, famotidine, and nizatidine. †Omeprazole, pantoprazole, and lansoprazole.

28.2±15.6

39.2±22.4 37.9±4.5

51.9±17.1

83.6±11.4

0

20

40

60

80

100

Pati

ents

Hea

led

(%

)

Placebo

Sucralfate

Cisapride

H2RAs*

PPIs†

FIGURE 1. Overall pooled healing rates for placebo, sucralfate,cisapride, H2RAs, and PPIs in patients with Savary-Miller grades IIto IV erosive esophagitis.23



practice, antacids help to control mild to moderatereflux symptoms in a large proportion of patients.26

Because they act locally, antacids are considered first-line therapy for pregnant women who experienceheartburn. However, magnesium-containing agentsshould be avoided in the latter part of pregnancy.19

Clinical efficacy. Despite widespread use ofantacids, definitive evidence of their therapeuticbenefit in the treatment of GERD is limited by thepaucity of well-designed, large, placebo-controlledtrials. For the placebo-controlled studies that areavailable, results are conflicting.

One placebo-controlled study comparing a high-dose antacid (10 mL seven times daily) with anH2RA in 37 patients with GERD found that symp-tom improvement was better in the antacid groupthan in the placebo group, but healing of erosiveesophagitis was not. However, another study in 32patients found that placebo actually performedslightly better than the high-dose antacid (15 mLseven times daily) in relieving GERD symptoms andin healing esophagitis.11

Many studies of the efficacy of antacids in com-bination with alginic acid have produced favorableresults in terms of GERD symptom relief. However,data from these studies, including a non–placebo-controlled comparative trial in children, two openstudies without placebo groups, a nonblinded study,and several comparative trials, are of limited usebecause of the lack of true placebo controls.25

Most studies testing the efficacy of antacids havefound that, even at high doses, their effect on heal-ing erosive esophagitis is no better than that of

placebo.11,26 For example, in a 4-week, randomized,double-blind, placebo-controlled trial, Furman andcolleagues27 compared a high-dose liquid antacid (7oz/day) given 15 minutes and 1 hour after mealswith cimetidine 300 mg four times daily. Patientsunderwent endoscopy, biopsy, and acid perfusiontesting at baseline and at the end of the study. Theyalso were asked to complete symptom diary cardsduring the study. Only patients given cimetidinehad a significant reduction in the frequency andseverity of heartburn (P < 0.05). The liquid antacidwas similar to placebo in its reduction of heartburnseverity. No treatment improved regurgitation.Furthermore, neither cimetidine nor the liquidantacid improved any objective measure of GERDseverity (endoscopy, acid perfusion test results).

Often in clinical trials, a reduction in the use ofantacids is a hallmark of efficacy for other agents(ie, H2RAs or PPIs).

■ HISTAMINE H2-RECEPTOR ANTAGONISTS

Histamine H2-receptor antagonists are acid-sup-pressive agents that treat GERD by decreasing acidsecretion and thus decreasing the causticity of therefluxate. The H2RAs approved for use in theUnited States are cimetidine, ranitidine, famoti-dine, and nizatidine. All are available by prescrip-tion and in over-the-counter formulations that areusually one half the standard prescription dose.

Although there are some variations in potency,duration, and onset of action, H2RAs have similarefficacy rates in symptom relief and healing ofesophagitis.1 Because they are all metabolized viathe cytochrome P450 system in the liver, some druginteractions can occur. However, as a class, they areconsidered very safe, with few side effects.28 In clin-ical trials, these agents consistently reduced GERDsymptoms and promoted esophageal healing at arate significantly better than placebo, especially inpatients with milder grades of esophagitis.

Pharmacokinetic/pharmacodynamic overviewOral absorption of H2RAs is fairly rapid, with peakplasma concentrations attained within 1 to 3 hoursafter administration. A second peak after oraladministration has been observed with all H2RAsexcept nizatidine. Mean oral bioavailability differssomewhat among the agents, ranging from a low of40% with famotidine to a high of 80% with cime-tidine.

100

80

60

40

20

0

100

80

60

40

20

0

Symptomatic Response Esophagitis Healing

PlaceboH2RAOmeprazole

27

60

(n = 1,887)

83

(n = 834)

24

(n = 433)

50

(n = 1,003)

78

(n = 754)

Pati

ents

(%

)

Pati

ents

(%

)

(n = 562)

FIGURE 2. Symptom response rates and esophagitis healingrates in patients with GERD receiving H2RAs, omeprazole, orplacebo. The H2RA dosages represented here are cimetidine 800 to1,600 mg/day, ranitidine 300 to 600 mg/day, nizatidine 600mg/day, and famotidine 40 mg/day, all given in divided doseseither twice or four times daily. Omeprazole was given at adosage of 20 mg to 60 mg once daily.1


M AT O N

Plasma concentrations of H2RAs and inhibitionof gastric acid secretion are directly related, imply-ing a rapid equilibration between drug concentra-tion in plasma and at the site of action.29

In general, the acid-suppressive abilities of H2RAsare more effective on nocturnal acid secretion. Dura-tion of acid inhibition is longer when the drug istaken in the evening or before bedtime. Equipotentdoses of H2RAs equally inhibit acid secretion.

Efficacy: symptom improvementOverall, H2RAs relieve symptoms in 60% ofpatients with GERD, whereas placebo relievessymptoms in 27% (Figure 2).1 H2RAs are effectivein the control of nocturnal acid reflux episodes aswell.11 In patients with no erosive esophagitis andwith mild or intermittent GERD symptoms, symp-tomatic treatment response rates are 70% or high-er.21 Patients with esophagitis experience lower ratesof symptom relief and are more likely to have symp-tom relapse once therapy is stopped.

Relapse. Hallerback and colleagues30 examinedsymptom relapse in patients after a 4-week course ofeither ranitidine or placebo. The study included 423patients with GERD symptoms, most of whom hadmild reflux disease; 67.4% had either a normal-appearing esophagus or erythema only. Another28.1% had grade I esophagitis, which consists ofsmall, isolated lesions. Only 4.5% of this study pop-ulation had esophagitis grades II through IV.Patients with more extensive injury were excluded.

Initially, patients were randomized to receiveeither ranitidine 150 mg twice daily or placebo for 2weeks. After the initial trial, those who were satis-

fied with their treatment (ie, responded to therapywith either improved or complete relief) continuedwith that therapy. Patients who were not satisfiedwith their treatment were then re-randomized toreceive ranitidine 150 mg two or four times daily foranother 2 weeks. Patients whose symptoms did notrespond after 4 weeks of therapy were removed fromfollow-up. After 4 weeks of therapy, all responderswere taken off therapy and followed for an addi-tional 24 weeks.30

Figure 3 shows the symptomatic relapse rates inthe total population and in patients with and with-out erosive esophagitis. At 24 weeks of follow-up,symptom relapse had occurred in 52% of patientswith GERD who did not have erosive esophagitiscompared with 67% of those who did have erosiveesophagitis (P = 0.013).30

Dosage level. In the Hallerback study,30 the per-centage of patients who experienced symptomimprovement or complete relief at week 4 was sim-ilar in all groups that received ranitidine, regardlessof dosage level. This finding was confirmed in a sim-ilarly designed study of longer duration. In thisstudy, Kahrilas and colleagues31 compared high-doseand standard-dose ranitidine therapy in patientswho remained unresponsive after 6 weeks of raniti-dine 150 mg twice daily. Of the 481 patients withGERD symptoms who initially received a 6-weekcourse of 150 mg twice daily, 285 (59%) remainedsymptomatic. Of this group, 270 were re-random-ized to receive either 150 mg or 300 mg of ranitidinetwice daily. After an additional 8 weeks of therapy,only 44.8% of patients receiving the higher dose ofranitidine and 45.4% of those receiving the lower

0

20

40

60

100

80

24 Weeks of Follow-Up

52

67†

59

Pati

ents

Wh

o E

xper

ien

ced

Rel

apse

(%

)* No Erosions at EntryErosions at EntryTotal Population

*Relapse defined as requiring medication for reflux symptoms.†P = 0.013 vs patients with no esophageal erosions at entry.

FIGURE 3. GERD symptom relapse rates after discontinuation ofa successful 4-week course of treatment with ranitidine 150 mgtwo or four times daily or placebo.30

Responders*Nonresponders

Pati

ents

(%

)

Pati

ents

(%

)

P = not statistically significant.

Ranitidine (150 mg bid)

Initial 6-Week Course(N = 481)



Additional 8-Week Course (N = 270†)

0

20

40

60

80

100

0

20

40

60

80

100

*Patients with no or mild heartburn.†Patients who did not respond to initial treatment course.

41

59

45.4

54.6

44.8

55.2

FIGURE 4. Rates of treatment response (heartburn relief) to aninitial course of standard-dose ranitidine therapy (150 mg twicedaily) and to an additional course of standard-dose or high-dose(300 mg twice daily) ranitidine therapy in patients with GERD.31



dose achieved treatment response (mild or no heart-burn) (Figure 4).

Efficacy: healing of erosive esophagitisIn general, the results of treatment with H2RAs arenot as good in patients with severe erosive esophagi-tis. However, healing of erosions occurs in 50% ofpatients with erosive esophagitis who are treatedwith H2RAs.26,28

Placebo-controlled studies demonstrate thatH2RAs provide better symptomatic relief of GERDand mucosal healing of erosive esophagitis than doesplacebo. Although many trials do not contain aplacebo arm, Tougas and Armstrong28 cited a place-bo healing rate of 16.5% at 12 weeks for patientswith grade II to grade IV esophagitis in a review ofthe efficacy of H2RAs for treatment of GERD. Thatcompared with a mucosal healing rate of 39.7% (P <0.0005) for H2RAs in the same review. Placebo wasassociated with a 13% rate of overall symptom relief,

compared with a 44% rate of overall relief withH2RAs (P < 0.001). All four H2RAs, when used atthe usual recommended dose, were equally safe andeffective, although their efficacy was limited in moresevere forms of GERD, such as erosive esophagitis(endoscopic healing rates of 40% to 50%, symptomimprovement rates of 40% to 60%).28

Some non–placebo-controlled comparative stud-ies involving H2RAs have produced 12-week heal-ing rates as high as 70%.25 For example, a compara-tive study of ranitidine (150 mg two or four timesdaily) and cimetidine (800 mg twice daily) for thehealing of erosive esophagitis demonstrated 12-weekhealing rates ranging from 68% to 77% (Figure 5).32

Another study of esophagitis healing found that the6-week healing rate with ranitidine was 78% for iso-lated erosions but dropped to 38% for confluent ero-sions and to only 23% for circumferential erosions(Figure 6).33

ToleranceThe suppression of intragastric acidity diminisheswith repeated administration of H2RAs.34 A singledose of an H2RA (Table 3)35 inhibits acid secretionfor approximately 4 to 8 hours and decreases stimu-lated acid secretion by approximately 70% inpatients with esophagitis. However, H2RA treat-ment has several disadvantages, including a rela-tively short duration of action (compared withPPIs), incomplete inhibition of acid secretion inresponse to a meal, and the development of toler-ance. It is therefore not easy for H2RAs to effec-tively heal the more severe forms of erosiveesophagitis, even when very high doses are used.When severe erosive esophagitis is not present,relief of reflux symptoms has been obtained after 4weeks using a twice-daily regimen. In grades I or II

TABLE 3H2RAs indicated for the treatment of GERD andtheir recommended nonprescription dosages35

H2RAs Dosage (over-the-counter)

Nizatidine 75 mg twice daily, as needed

Famotidine 10 mg twice daily, as needed

Cimetidine 200 mg twice daily, as needed

Ranitidine 75 mg twice daily, as needed

0

20

40

60

80

Isolated Erosions

Confluent Erosions

Circumferential Erosions

Ero

sive

Eso

ph

agit

is H

ealin

g R

ate

(%) 100

78

38

23

6 WeeksN = 108.

FIGURE 5. Comparison of erosive esophagitis healing rates withranitidine 150 mg two or four times daily and cimetidine 800 mgtwice daily.32

FIGURE 6. Healing of erosive esophagitis with ranitidine 150 mgor 300 mg twice daily, by severity of disease.33

0

20

40

60

80

Week 4 Week 8 Week 12

Ranitidine 150 mg bid (n = 236)Ranitidine 150 mg qid (n = 229)Cimetidine 800 mg bid (n = 231)

100

Pati

ents

Wit

h H

eale

d E

rosi

ve E

sop

hag

itis

(%

)

*P<0.02 vs cimetidine.

38

49*

37

56

67

52

71

77*

68


M AT O N

esophagitis, healing can be achieved with any of theH2RAs in 40% to 60% of cases in 8 weeks. This suc-cess rate can be increased to 50% to 70% by a sub-stantial increase in dose, but this entails cost andcompliance considerations.36

It is possible that tolerance may develop as aresult of the down-regulation of H2-receptors.Another possibility is that there are adaptivechanges in acid secretion that are stimulated byacetylcholine, gastrin, or both.

A number of clinical studies have shown that tol-erance to standard H2RAs probably develops with-in the first 2 weeks of therapy.36 More recently, how-ever, tolerance has been shown to develop within72 hours when intravenous administration of raniti-dine is used to control bleeding in the upper gas-trointestinal tract. To avoid such tolerance, morefrequent dosing of ranitidine, more careful monitor-ing of intragastric pH, and repeated dose adjust-ments would be needed.37

■ PROTON PUMP INHIBITORS

Proton pump inhibitors are the most effective med-ical treatment for GERD. They profoundly suppressacid secretion through inhibition of H+,K+ adeno-sine triphosphatase (ATPase), the proton pump ofthe parietal cell responsible for acid production.Unlike H2RAs, they block acid production regard-less of the method of stimulation, providing agreater degree of acid suppression for a longer dura-tion of time. All PPIs are prodrugs, so-called substi-tuted benzimidazoles, which must be activated byacid to inhibit the proton pump. This translates intohigher efficacy rates in terms of GERD symptomrelief and healing of erosive esophagitis.26 A once-daily, morning dose of a PPI will relieve symptomsin 83% of patients with GERD and heal erosiveesophagitis in 78%.1 Furthermore, these rates areachieved after only 4 to 8 weeks of therapy. As withH2RAs, healing of esophagitis with PPIs correlateswith the severity of esophagitis.

Excellent healing rates have been reported ineven the most severe grades of esophagitis after PPItherapy.26 There is a wealth of study data on thesafety and efficacy of omeprazole, the first PPIapproved for treatment of GERD. Other availablePPIs include lansoprazole, rabeprazole, and pantop-razole. Clinical efficacy in GERD and the safetyprofiles among this first generation of PPIs are verysimilar. The newest PPI, esomeprazole, the S-isomer

of omeprazole, has demonstrated more completesymptom relief in patients with GERD andesophageal healing for a greater proportion ofpatients and in a shorter time period compared withomeprazole.38,39

Pharmacokinetics and pharmacodynamicsAll PPIs are metabolized in the liver via the cyto-chrome P450 system, specifically by the CYP2C19and CYP3A4 enzymes.40 There are subtle differ-ences, however, in how each PPI is metabolizedwithin this system. An agent’s preference for oneenzyme over another influences the metabolic path-way and leads to differences among the PPIs ininteractions with other drugs (Table 4).

Because PPIs reduce gastric acidity, they mayalter the absorption of other orally administereddrugs. Elevated gastric pH has the potential to affectthe stability of agents that are acid-labile or alka-line-labile, as well as the absorption of agents thathave pH-dependent formulations. Whether theseinteractions are clinically significant is more diffi-cult to determine. However, PPIs reduce the area

TABLE 4Possible drug interactions with PPIs resulting frommetabolism via the cytochrome P450 system40–43

PPI Interactions

Omeprazole • Inhibits metabolism of phenytoin,diazepam, antipyrine, aminopyrine,and the R-isomer of warfarin

• Does not inhibit metabolism of propranolol, theophylline, or the s-isomer of warfarin

Lansoprazole • No clinically significant interactionswith most drugs metabolized throughthe cytochrome P450 system

• Increases metabolism rate of theophylline by 10%

Pantoprazole • No clinically significant interactionswith drugs metabolized through thecytochrome P450 system

• No interactions with oral contraceptives

Rabeprazole • No clinically significant interactionswith drugs metabolized through thecytochrome P450 system

Esomeprazole • No clinically significant interactionswith drugs metabolized through thecytochrome P450 system



under the curve (AUC) and the peak plasma levelsof ketoconazole to such an extent that patientsstarting therapy with ketoconazole may need to dis-continue PPI therapy.40

Individual PPIs differ in clinical pharmacology.The following sections identify selected features ofthe clinical pharmacology of the five available PPIs.

Omeprazole is inactivated when it is exposed togastric acid, so it has been formulated in granulesthat release the drug only when the pH is greaterthan 6.0, resulting in a bioavailability of approxi-mately 50%. Peak plasma concentrations occur 2 to4 hours after oral administration and increase dur-ing the first days of therapy. The plasma half-life ofomeprazole is approximately 1 hour. Food intakehas no effect on the drug’s pharmacokinetics.40

Omeprazole differs from other PPIs in that itsbioavailability increases with repeat dosing. A sin-gle dose of omeprazole has a bioavailability ofapproximately 35%, which increases with repeatdosing to around 60%. Mean AUC and peak plasmaconcentrations also increase disproportionatelybetween days 1 and 5 of treatment.40

One 20-mg dose of omeprazole inhibits acidsecretion by 65% after 4 to 6 hours, dropping to 25%after 24 hours. Inhibition increases after subsequentdoses and plateaus after four to six doses. Steady-state inhibition varies widely among patients, rang-ing from 35% to 65% based on acid secretion mea-surements taken 24 hours after drug inhibition, andfrom 30% to 100% based on measurements of 24-hour gastric acidity based on intragastric pH.41 Inone study, 20-mg and 40-mg doses resulted in amean intragastric pH greater than 4.0 for approxi-mately 42% and 62%, respectively, of a 24-hourperiod after 5 days of administration.42 In general,when larger doses of omeprazole are given, variationof acid inhibition in patients is reduced, and acidinhibition is increased.41

Lansoprazole. Like other PPIs, lansoprazole isacid-labile. The drug is rapidly absorbed, with peakplasma concentration reached only 1.7 hours afteradministration. Food intake with lansoprazoledelays the drug’s absorption.40 Multiple dosing doesnot alter its pharmacokinetics. Lansoprazole’s bio-availability is approximately 80%, and its elimina-tion half-life is less than 2 hours.43

Some, but not all, studies have shown that lan-soprazole’s ability to suppress acid secretion is dose-dependent and increases with repeated administra-tion. In clinical trials, a 30-mg dose resulted in an

intragastric pH greater than 4.0 for 41% of a 24-hourperiod on the first day of administration, rising to66% on the fifth day. A 15-mg dose yielded an intra-gastric pH greater than 4.0 for 22% of a 24-hourperiod on the first day of administration and for 49%of a 24-hour period by the fifth day. The onset ofantisecretory activity also differs between doses, withthe 30-mg dose causing an increase in intragastricpH in 1 to 2 hours after administration and the 15-mg dose causing an increase in 2 to 3 hours. Thistime decreased with repeated dosing for both dosagestrengths.43

Pantoprazole. All PPIs are rapidly degraded byacidic conditions, but pantoprazole is slightly morestable than omeprazole and lansoprazole under neu-tral conditions or conditions that are mildly acidic(pH ≈3.5 to pH ≈7.4). Pantoprazole is therefore eas-ier to produce in intravenous form.42 A 40-mg dose ofpantoprazole reaches peak plasma concentrations 2to 4 hours after administration. The estimated bio-availability of 77% reflects a low first-pass hepaticextraction. Food intake delays absorption of pantop-razole. The drug’s mean plasma terminal eliminationhalf-life is 0.9 to 1.9 hours. Repeated dosing does notalter the pharmacokinetics of pantoprazole.42 In thisway, pantoprazole is similar to lansoprazole but unlikeomeprazole.

Steady-state acid inhibition by a repeated once-daily dose of pantoprazole is dose-related over therange of 20 mg to 60 mg. However, minimal addi-tional inhibition occurs with higher doses. After 5days of oral administration of 40 mg or 60 mg, sig-nificant reductions in basal, nocturnal, and 24-hourintragastric pH occurred. The 40-mg dose achievedan intragastric pH greater than 4.0 for approximate-ly 41% of a 24-hour cycle after 5 days of adminis-tration.42

Rabeprazole is degraded by acid in a manner sim-ilar to lansoprazole and omeprazole and is less stableat a neutral pH than other PPIs. Maximum plasmaconcentrations occur 3 to 4 hours after a single dose,regardless of the dosage strength. The bioavail-ability of rabeprazole following a 20-mg single doseis approximately 52%. Peak plasma concentrationsand the AUC increase with rising dosages, but thepharmacokinetics are not altered by multiple doses.Rabeprazole’s plasma half-life is similar to that ofomeprazole at approximately 1 hour. The time toachieve maximum plasma concentration is signifi-cantly prolonged by food intake. This is of no clini-cal significance, however, as the AUC is not altered


M AT O N

to a significant extent. It has been suggested thatthe effect of food on the time to maximum plasmaconcentration is secondary to the effect of food inprolonging the gastric emptying time of the enteric-coated tablets.40

The acid-suppressive effects of rabeprazole do notincrease to a significant extent with increasing doses.For instance, patients with GERD achieved an intra-gastric pH of 4.0 or greater for approximately 16 hoursafter 5 days of administration of 20 mg of rabeprazole,compared with approximately 18 hours after 5 daysof administration of 40 mg.44

Esomeprazole. Discussion of esomeprazole requiresbrief mention of stereoisomers, which are moleculeswith one or more “chiral” centers that allow the pos-sibility of forms with the same chemical formula butdiffering spatial arrangements. These differences cantranslate into clinical differences in terms of a com-pound’s efficacy and toxicity.

Esomeprazole is the S-enantiomer of racemic S,R-omeprazole. It has been shown to be stable, withmore than 40% of each dose showing conversion tothe R-isomer.45

Esomeprazole reaches maximum plasma concen-tration approximately 2 hours after administration ofa single dose. As with omeprazole, the bioavailabili-ty of esomeprazole is altered by repeat dosing. In clin-ical studies, a 20-mg dose of esomeprazole was 50%bioavailable on day 1 of administration, increasing to68% on day 5. The total AUC increased by 90%over 5 days. With a 40-mg dose of esomeprazole,bioavailability increased from 64% on day 1 to 89%on day 5, with an AUC increase of 159%.46 The plas-ma clearance rate of esomeprazole decreases from 22L/hour to 16 L/hour over a 5-day period, and its plas-ma elimination half-life increases from 0.8 hours to1.2 hours with repeat dosing.

The pharmacokinetic and pharmacodynamicprofile of esomeprazole differs from that of omepra-zole. Esomeprazole given daily for 5 days had a 70%higher AUC than the same dosage of omeprazolegiven over the same period.47 Furthermore, althoughthere is variability in omeprazole’s acid inhibitionamong individual patients, this effect is substantial-ly reduced with esomeprazole.

Efficacy: symptom improvementSymptom relief is the primary goal of medical ther-apy for GERD and is highly predictive of endoscop-ic healing if esophagitis is present.48 Interestingly,the severity of a patient’s heartburn is not necessar-

ily predictive of the severity of erosive esophagitisthat he or she experiences. This was clearly shownin a clinical trial by Venables and colleagues,49 whofound that chronic GERD symptoms were unreli-able in predicting the presence of underlyingesophagitis in patients at trial entry. Althoughpatients with severe heartburn may have normalendoscopy findings, many will have undetected ero-sive esophagitis. This study compared the efficacy ofomeprazole 10 mg or 20 mg daily with that of ranit-idine 150 mg twice daily for relief of heartburn. The20-mg omeprazole dose was the most effective ini-tial therapy for relief of GERD symptoms.49

Many studies have compared the effects ofH2RAs and PPIs on symptom improvement. Ingeneral, PPIs are more effective than H2RAs andwork at a faster rate. Additionally, omeprazole hasbeen compared with other PPIs for symptomimprovement in several trials. Overall, there is nota significant difference in efficacy among the PPIs;however, there is some variability in different clin-ical trials.

In their meta-analysis of the efficacy of GERDtherapies for treating grades II through IV erosiveesophagitis, Chiba and colleagues23 showed that PPIsprovide complete heartburn relief in a higher per-centage of patients and at a faster rate comparedwith H2RAs. They found that more patients becameheartburn-free by the second week of treatment withPPIs (58.0% ± 16.9%) than by 8 weeks of therapywith H2RAs (48.8% ± 16.2%). The speed of heart-burn relief was faster with PPIs than with H2RAs: atweek 2, patients treated with PPIs became heart-burn-free at a rate of 31.8% (±7.9%) per week, com-pared with a rate of 17.9% (± 5.8%) per week forpatients treated with H2RAs (Figure 7).

In the same analysis, PPIs also provided thegreatest overall symptom relief, as 77.4% (±10.4%)of PPI-treated patients became heartburn-free, com-pared with 47.6% (±15.5%) of patients treated withH2RAs.23

In a pair of randomized, double-blind, multicentertrials, Richter and colleagues50 compared lansopra-zole with ranitidine in 901 patients with sympto-matic reflux disease confirmed by endoscopy to benonerosive GERD. The frequency of antacid useserved as an end point for evaluating these agents’efficacy in relieving heartburn symptoms. Comparedwith patients who received either of two dosages oflansoprazole, patients treated with ranitidine report-ed ingesting antacids on a significantly higher per-



centage of days (Figure 8) and ingesting a signifi-cantly higher number of antacid tablets per day.Across all treatment groups, the frequency of antaciduse was associated with the frequency of days ornights with heartburn.

Variations among PPIs. Although all PPIs pro-vide a comparably high level of symptom improve-ment, there are some variations in PPI performancein comparative clinical trials. Castell and col-leagues51 evaluated two doses of lansoprazole (15 mgand 30 mg once daily) in comparison with omepra-zole (20 mg once daily) and placebo in 1,284 patientswith endoscopically confirmed erosive esophagitis.

All PPI-treated groups experienced relief of daytimeand nighttime heartburn, regurgitation, and belch-ing to a significantly greater degree than did theplacebo group, as judged by investigators andpatients. Omeprazole and the 30-mg dose of lan-soprazole were more effective than placebo in reliev-ing investigator-assessed painful swallowing, whereasthe 15-mg lansoprazole dose was not significantlymore effective than placebo. There were no differ-ences between the omeprazole and lansoprazole 30-mg treatment groups in investigator-assessed symp-tom relief, but there were differences in patient-assessed relief. After 8 weeks of therapy, patients whoreceived omeprazole reported experiencing heart-burn on 11.8% of days and 8.9% of nights during thetrial, whereas those receiving 30 mg of lansoprazolereported experiencing heartburn on 8.6% of days and6.5% of nights (P < 0.05 vs omeprazole). Patientsreceiving placebo reported experiencing heartburnon 60% of days and 45% of nights.

Mössner and colleagues52 investigated symptomrelief in 286 patients with grades II or III erosiveesophagitis randomized to receive 40 mg of pantop-razole or 20 mg of omeprazole daily for 8 weeks.Investigator-assessed symptom relief was recordedafter 2 weeks and 4 weeks of therapy. Differencesbetween the treatment groups in relieving heartburn,regurgitation, and painful swallowing were not statis-tically significant at 2 weeks or 4 weeks. At 2 weeks,59% of patients receiving pantoprazole and 69% ofthose receiving omeprazole were symptom-free; at 4weeks, these percentages rose to 83% and 86%,respectively (not statistically significant).

Dekkers and colleagues53 compared the efficacy of

FIGURE 8. Four-week and 8-week comparisons of the medianpercentage of days antacids were used by patients with nonero-sive GERD while receiving either ranitidine or one of two dosagesof lansoprazole for relief of heartburn symptoms.50

*P = 0.001 vs ranitidine. †P = 0.01 vs ranitidine.

32.128.5

21.4 19.6†16.7*17.9*

0

20

40

60

80

100

Med

ian

Day

s o

f A

nta

cid

Use

(%

)


Lansoprazole (15 mg qd)Lansoprazole (30 mg qd)

4 Weeks 8 Weeks

FIGURE 7. Time curves for rates of heartburn symptom relief (top)and esophagitis healing (bottom) with PPI and H2RA therapy in ameta-analysis of efficacy trials among patients with GERD.23 Byevaluation week 2, PPIs relieved heartburn and healed esophagitisin more patients than H2RAs did by weeks 8 or 12. Reprinted fromreference 23 with permission from the American GastroenterologicalAssociation.

80

PPIsH2RAsPlacebo

60

40

20

0

80

100

60

40

20

0

Weeks1-2

Weeks3-4

Weeks6-8

Week2

Pati

ents

Hea

rtb

urn

-Fre

e (%

)

Symptom Relief

Esophagitis Healing

Pati

ents

Hea

led

(%

)

Week4

Week6

Week8

Week12


M AT O N

omeprazole 20 mg daily and rabeprazole 20 mg dailyin an 8-week trial in 202 patients with erosive orulcerative reflux disease. Similar improvements inheartburn frequency rates and in daytime and night-time heartburn severity were seen in the two treat-ment arms. At 8 weeks, 73% of patients receivingrabeprazole and 76% of those receiving omeprazolereported a lessening of heartburn frequency. In termsof heartburn severity, 68% of patients receivingrabeprazole reported resolution of daytime heartburnand 64% reported resolution of nighttime heartburn,which were comparable to the rates in the omepra-zole group (66% and 67%, respectively).

Kahrilas and colleagues38 compared symptomrelief with omeprazole 20 mg daily and esomeprazole20 mg or 40 mg daily in 1,960 patients with erosiveesophagitis. After 4 weeks of therapy, esomeprazole40 mg daily provided more effective relief of symp-toms: patients who received this regimen reportedexperiencing no heartburn on 72.7% of days and84.7% of nights during the trial, whereas thosereceiving omeprazole reported experiencing noheartburn on 67.1% of days and 80.1% of nights (P< 0.05). Onset of symptom relief was also faster withesomeprazole 40 mg daily, as 46.6% of patientsreceiving this regimen reported no heartburn on thefirst day of treatment, compared with 37.0% ofpatients receiving omeprazole.

Finally, a study by Richter and colleagues of2,425 patients with GERD demonstrated betterresolution of investigator-assessed heartburn andregurgitation after 4 weeks of treatment withesomeprazole 40 mg daily than with omeprazole 20mg daily (Figure 9).39,54

Symptom relief in nonerosive GERD. PPIs havealso been tested in patients who have GERD symp-

toms but do not have erosive esophagitis. Lind andcolleagues55 compared omeprazole 10 mg or 20 mgdaily with placebo in 509 patients with GERD butwithout erosive esophagitis. After 4 weeks of thera-py, 46% of patients receiving omeprazole 20 mgreported complete absence of heartburn, comparedwith 31% of patients receiving omeprazole 10 mgand 13% of placebo recipients (Figure 10). A studyby Bate and colleagues56 of 209 patients with GERDsymptoms but no esophagitis yielded similar results.This study tested only the 20-mg dose of omeprazoleagainst placebo. After 4 weeks of therapy, 57% ofpatients in the omeprazole group were free of heart-burn (vs 19% in the placebo group [Figure 10]),75% experienced no regurgitation (vs 47% withplacebo), and 43% were completely asymptomatic(vs 14% with placebo).

While PPIs are significantly more effective thanplacebo in relieving heartburn in patients with ero-sive esophagitis, they are not as effective in patientswith nonerosive reflux disease. Carlsson and col-leagues57 conducted a 4-week comparison of twodosages of omeprazole in 277 patients with erosiveesophagitis and 261 patients with GERD symptomswithout erosive esophagitis. Omeprazole 10 mg dailyachieved complete symptom relief in 37% of patientswith erosive esophagitis, compared with 31% ofpatients without erosive esophagitis. Similarly,omeprazole 20 mg daily achieved complete symptomrelief in 48% of patients with erosive esophagitis vs29% of those without erosive esophagitis.

Dosage level. Dosage is an important considera-tion in PPI therapy. In the absence of esophagitis,when symptoms are mild or intermittent, a standard-dose PPI has been found to be effective.57–59 If symp-toms are particularly troublesome or there is moder-

Study byLind and Colleagues

(n = 509)

Study byBate and Colleagues

(n = 209)

0

20

40

60

80

100

*P<0.0001.

13

46*

19

57*

PlaceboOmeprazole (20 mg qd)

Pati

ents

Hea

rtb

urn

-Fre

e at

4 W

eeks

(%

)

FIGURE 10. Percentages of patients with GERD but without ero-sive esophagitis who experienced complete relief of heartburnafter 4 weeks of PPI therapy or placebo treatment.55,56

Heartburn Regurgitation

Pati

ents

(%

)

0

20

40

60

80

100Omeprazole 20 mg (n = 1,209)Esomeprazole 40 mg (n = 1,216)

58.1

68.3*75.2

80.1†

*P<0.001. †P<0.003.

FIGURE 9. Percentages of patients with investigator-assessedresolution of GERD symptoms after 4 weeks of omeprazole oresomeprazole therapy.39,54



ate or severe erosive esophagitis, a twice-daily dosagemay be necessary for a period of time. (However, noprospective data exist to support this recommenda-tion.60) Following this, dose reduction should beattempted and a plan formulated for long-term ther-apy. Dosage level is significant to controlling symp-toms and maximizing the success of PPI therapy.60

Nocturnal acid breakthrough. Several studieshave assessed various therapeutic regimens for con-trolling the persistent problem of nocturnal acidbreakthrough, including double-dose PPI therapyand single- or double-dose PPI therapy combinedwith an H2RA. Khoury and colleagues61 found thatcombination therapy with 20 mg of omeprazole inthe morning and 150 mg of ranitidine at night is notas effective in controlling intragastric pH as omepra-zole 20 mg twice daily (one dose in the morning andone at night). In 20 healthy volunteers, the medianpercentage of time that intragastric pH was less than4 when participants were upright was 29.7% in theranitidine group and 18.9% in the double-doseomeprazole group (P = 0.003). The median percent-ages of time that intragastric pH was less than 4when participants were recumbent were 44.75% and23.45%, respectively (P = 0.02). In this study, rani-tidine administered at bedtime did not eliminate theneed for a second dose of omeprazole.

Peghini and colleagues62 compared three differ-ent regimens for controlling nocturnal acid break-through in 12 healthy volunteers. All participantsreceived omeprazole 20 mg twice daily. They alsoreceived either an additional dose of omeprazole,150 or 300 mg of ranitidine, or placebo at bedtime.Participants who received placebo at bedtime expe-rienced an intragastric pH of less than 4 for 48% ofthe night. A third dose of omeprazole reduced thispercentage to 31% (P < 0.005), but ranitidine wasthe most effective therapy: participants whoreceived 150 or 300 mg of ranitidine at bedtimeexperienced an intragastric pH of less than 4 foronly 5% and 6% of the night, respectively (P < 0.01vs omeprazole 20 mg three times daily). These datasuggest that double-dose PPI therapy daily plus anH2RA at bedtime may be an effective regimen forcontrol of nocturnal acid breakthrough.

Efficacy: healing of erosive esophagitisIn their above-mentioned meta-analysis (Figure 1),Chiba and colleagues23 noted that PPIs were moreeffective than H2RAs in treating patients withgrades II through IV erosive esophagitis. PPIs pro-

vided more complete relief of symptoms and fasterhealing of esophagitis (Figure 7). At week 2, thehealing rate per week was 31.7% (±3.3%) for PPIs,compared with 15.0% for H2RAs. The healing rateper week slowed for both agents at each subsequent2-week interval. However, PPIs maintained a thera-peutic advantage because more patients were healedearlier in the course of PPI therapy, leaving fewerpatients who were available to heal in later weeks.Overall, PPIs produced healing at a rate of 11.7% (±0.5%) per week, which was twice as fast as the heal-ing rate per week with H2RAs (5.9% ± 0.2%) andfour times as fast as that with placebo (2.9% ±0.2%). A mean of 83.6% (±11.4%) of patients witherosive esophagitis were healed with PPI therapy,compared with 51.9% (±17.1%) with H2RA thera-py. These numbers compare with a healing rate of28.2% (±15.6%) with placebo.

A meta-analysis by Caro and colleagues63 alsofound that PPIs were significantly more effectivethan H2RAs. The investigators analyzed 53 random-ized controlled trials, 38 of which involved acutetherapy, although 12 of these were subsequentlyexcluded and 15 involved maintenance treatment.Of the 26 acute therapy trials, 18 compared a PPIwith an H2RA. Of the 15 maintenance therapy tri-als, 5 compared a PPI with an H2RA. No study ofpantoprazole met the inclusion criteria for mainte-nance therapy. Combined efficacy rates from bothacute and maintenance studies yielded a risk ratiothat was highly favorable to PPIs.

Variations among PPIs. In the same analysis, Caroand colleagues63 compared the efficacy of lansopra-zole, pantoprazole, and rabeprazole with that ofomeprazole and found no differences in those head-to-head comparisons. Similar results were obtained byother investigators, including teams led by Castell,51

Mössner,52 and Dekkers.53

Although the study by Castell and colleagues51

found differences in patient-assessed symptom reliefbetween groups receiving omeprazole 20 mg daily orlansoprazole 30 mg daily, esophageal healing rateswere not statistically different between the groups.Eight-week courses of each regimen healed esopha-gitis in approximately 91% of patients. In the studyby Mössner and colleagues,52 patients with esopha-gitis received an 8-week course of either pantoprazole40 mg daily or omeprazole 20 mg daily. The healingrate with pantoprazole (94%) was not statistically dif-ferent from that with omeprazole (90%). Finally,Dekkers and colleagues53 found that an 8-week course


M AT O N

of rabeprazole 20 mg daily healed 92% of patientswith erosive esophagitis, compared with 94% for an8-week course of omeprazole 20 mg daily.

In the Castell study,51 lansoprazole 15 mg dailyhealed 79% of patients, a significantly lower per-centage than those achieved with lansoprazole 30mg daily and omeprazole 20 mg daily (P < 0.05). Asimilar dose-dependent pattern occurred in a studyby Lundell and colleagues64 between 10-mg and 20-mg doses of omeprazole. These researchers analyzedthe efficacy of omeprazole in the healing of LosAngeles (LA) classification grades A through Cerosive esophagitis by grade (mild through moder-ate-to-severe injury). With omeprazole 10 mg daily,healing efficacy was directly correlated to the gradeof esophagitis: 77% of patients with grade A werehealed, 50% of patients with grade B, and 20% ofpatients with grade C. However, this gradation inhealing did not occur with omeprazole 20 mg daily,which healed roughly the same percentage (approx-imately 80%) of patients with grades A and Besophagitis. The 20-mg dose of omeprazole healedfewer patients with grade C esophagitis (approxi-mately 40%) than with other grades, although italso healed more patients with this severity gradethan did the 10-mg dose. A dose-related increase inhealing efficacy with omeprazole was not observed,however, above the 20-mg dose.

Sontag and colleagues65 compared 20-mg and 40-mg daily doses of omeprazole with placebo in 230patients with GERD symptoms and erosive esopha-gitis. Whereas both doses of omeprazole were supe-rior to placebo in all measures, symptom relief andesophagitis healing rates were similar in the twoomeprazole groups. By the eighth week, 73.5% ofpatients receiving omeprazole 20 mg had completeesophageal healing, compared with 74.7% ofpatients receiving omeprazole 40 mg and 14.0% ofplacebo recipients. Omeprazole 20 mg achievedcomplete relief of daytime heartburn in 79.5% ofpatients and complete relief of nighttime heartburnin 79.5% of patients, compared with 81.6% and85.1%, respectively, for omeprazole 40 mg, and37.2% and 34.9%, respectively, for placebo. Al-though the higher dose of omeprazole resulted infaster relief of symptoms, differences between thetreatment arms were not statistically significant.

A recent meta-analysis by Edwards and col-leagues66 found that esomeprazole 40 mg daily pro-duced higher healing rates than omeprazole 20 mgdaily at 4 and 8 weeks after treatment (Figure 11).

In their analysis of 12 randomized controlled trials,these investigators found no significant differencesamong omeprazole and lansoprazole, pantoprazole,or rabeprazole at 4 and 8 weeks. They postulated thatthe superiority of esomeprazole to omeprazole maybe related to its more effective suppression of intra-gastric acid. Recent acid-suppression studies67 showthat esomeprazole maintains intragastric pH above 4significantly longer than lansoprazole, pantoprazole,or rabeprazole do, which may account for the differ-ences in healing among the agents observed in thismeta-analysis.

One of the studies included in the meta-analysis,conducted by Kahrilas and colleagues38 (described in“Efficacy: symptom improvement” above), com-pared healing rates among omeprazole 20 mg dailyand esomeprazole 20 mg and 40 mg daily. After 8weeks of therapy, esomeprazole 40 mg healedesophagitis in 94.1% of patients, whereas omepra-zole did so in 86.9% of patients (P < 0.05).

The study by Richter and colleagues39 in 2,425patients with erosive esophagitis confirmed thesefindings. After 8 weeks of therapy, erosive esophagitiswas healed in 93.7% of patients receiving esomepra-zole, compared with 84.2% of patients receivingomeprazole (by intention-to-treat analysis). Esomep-razole was more effective than omeprazole in healing

0.5 0.7 1.0 1.5 2.0

Relative Risk

Relative Risk Favors Other PPIs

FavorsOmeprazole

Lansoprazole

Pantoprazole

Rabeprazole

Esomeprazole

FIGURE 11. Relative risk of endoscopic healing at 8 weeks forall standard-dose PPIs compared with omeprazole 20 mg.Rectangles denote relative risk values from individual comparativetrials with omeprazole; horizontal lines running through rectanglesdenote 95% confidence intervals. Diamonds indicate fixed effectsfrom all comparative trials of the given PPI. Reprinted from refer-ence 66 with permission from Blackwell Publishing.



all grades of esophagitis, as shown in Figure 12.Among patients with LA grades C and D esophagitis,esomeprazole healed 85.8% of patients after 8 weeksof therapy, whereas omeprazole healed 68.1%.

More recently, a study by Castell and colleagues68

showed that esomeprazole demonstrated a slightlybut significantly higher healing rate (92.6%) thanlansoprazole (88.8%) at week 8. The difference inhealing rates between esomeprazole and lansoprazoleincreased as the baseline severity of erosive esophagi-tis increased.

Maintenance of healing. Lauristen and col-leagues69 demonstrated that esomeprazole was moreeffective than lansoprazole in maintaining the heal-ing of all grades of esophagitis (Figure 13). Theycompared esomeprazole 20 mg once daily with lan-soprazole 15 mg once daily in the maintenance treat-ment of 1,231 patients with healed reflux esophagi-tis. Analysis of remission rates based on the LA clas-sification system showed that esomeprazole main-tained patients in remission more consistently acrossall grades of reflux esophagitis, whereas the efficacyof lansoprazole decreased to a greater extent withincreasing severity of disease.

Other studies have demonstrated the low relapserate with esomeprazole therapy over a 6-monthperiod. In a study conducted by Johnson and col-leagues70 of 318 patients with erosive esophagitis,40-mg and 20-mg of doses of esomeprazole oncedaily were highly effective at maintaining healing oferosive esophagitis over 6 months. Rates of erosiveesophagitis recurrence were 6% and 7% with eso-meprazole 40 mg and 20 mg, respectively, compared

with 71% with placebo. Also, more than 70% ofpatients remained symptom-free at 6 months.

Table 5 presents the common dosing regimens andthe efficacy rates of the available PPIs for the main-tenance of erosive esophagitis healing.70–75

Intravenous PPI therapyIntravenous pantoprazole is the only PPI currentlyindicated in the United States for short-term treat-ment (7 to 10 days) of GERD in patients with a his-tory of erosive esophagitis who are unable to takethe oral formulation. In a double-blind placebo-controlled study, IV and oral pantoprazole wereshown to be similar in their ability to suppress max-imum and basal acid output. The study involved 65patients with erosive esophagitis who were given 20mg or 40 mg of oral pantoprazole for 10 days andthen randomized to receive either IV pantoprazoleor placebo for 7 days. Acid output was determined24 hours after the last day of oral medication and onthe first and last days of IV administration. Amongpatients receiving IV pantoprazole, acid suppressionwas comparable to that seen with oral pantoprazoleand was significantly better than that achieved bypatients receiving IV placebo. The recommendedadult dosage of IV pantoprazole is 40 mg daily for 7to 10 days.76

Omeprazole has been used in injectable form insome studies around the world in an effort to pre-vent rebleeding following treatment for bleedingulcers. A study in Hong Kong by Lau and col-leagues37 demonstrated the efficacy of omeprazole(given as an 80-mg bolus injection followed by a

Esomeprazole 20 mg qdLansoprazole 15 mg qd

Pati

ents

in R

emis

sio

n (

%)

*P<0.01.

Mild Disease Severe Disease

0

20

40

60

80

100

77

59

85*76*

0 30 60 90 120 150 180Days in Study Days in Study

Pati

ents

in R

emis

sio

n (

%)

0

20

40

60

80

100

0 30 60 90 120 150 180

FIGURE 13. Comparison of the maintenance of erosive esopha-gitis healing with esomeprazole and lansoprazole therapy inpatients with mild and severe disease. Reprinted from reference69 with permission from Blackwell Publishing.

81.3

70.463.8

90.4

80.0*87.2*89.4*

93.4*

0

20

40

60

80

100

Grade A Grade B Grade C Grade D

Week 8

Pati

ents

Hea

led

(%

)

Omeprazole 20 mg Esomeprazole 40 mg

*P<0.01 Cochran-Mantel-Haenszel.

(n = 813) (n = 972) (n = 497) (n = 140)

FIGURE 12. Grade-specific erosive esophagitis healing rates withomeprazole and esomeprazole therapy. Grades A and B in the LosAngeles (LA) classification represent mild to moderate esophagitis;grades C and D represent moderate to severe esophagitis.39


M AT O N

continuous infusion of 8 mg/hour for 72 hours) inpreventing recurrent bleeding after endoscopictreatment of bleeding peptic ulcers. This studyshowed that a high-dose infusion of omeprazolereduced the rate of recurrent bleeding, decreasedthe need for endoscopic treatment and blood trans-fusions, and shortened the length of hospitalization.

Failure of therapyTreatment of GERD with PPIs provides fast andcomplete relief to a larger percentage of patientsthan any other medical therapy. As well as PPIswork, however, they do not cure GERD and may notrelieve symptoms or heal esophagitis in all patients.

Treatment with a PPI may fail in some patientsbecause their symptoms are not caused by GERD.PPI therapy may also fail in a patient who does haveGERD if the PPI dosage or the duration of therapyis insufficient to control symptoms or heal esophagi-tis. Therapy may also be unsuccessful if the PPI failsto control gastric acidity, although such cases arerare.77,78 Several reasons have been postulated forwhy a PPI may fail to control gastric acidity. Theseinclude oral bioavailability differences, meal timingthat influences ATPase activation, increased metab-olism of the PPI by the cytochrome P450 system,and hypersecretion of acid (including Zollinger-Ellison syndrome).

Patients with higher grades of esophagitis andmore severe disease are less likely to experience com-plete healing with PPIs. A study conducted byHolloway and colleagues79 in 61 patients with gradesIII or IV erosive esophagitis found that 30% ofpatients were not healed after 8 weeks of therapywith omeprazole 20 mg daily. These patients hadgreater total 24-hour esophageal acid exposure beforetreatment than those whose esophagitis was healed.Of those patients in whom the original course oftherapy failed, 47% did not heal after 8 more weeksof therapy with a 40-mg daily dose of omeprazole.This final group of patients in whom both courses oftherapy failed had levels of acid exposure beforetreatment that were similar to those of patients whowere healed, but they had greater acid exposure dur-ing therapy, particularly at night while sleeping.

Several studies have been conducted withomeprazole to investigate the incidence of failure ofPPI therapy. Leite and colleagues80 studied 88patients with refractory GERD symptoms whoreceived 20 mg of omeprazole twice daily—twicethe usual dose. Twenty-four-hour pH monitoring

was used to assess results. Of the 88 patients, 17 hadan intragastric pH less than 4 for more than 50% ofa 24-hour period (considered failure of therapy).These 17 were then compared with 19 of the origi-nal 88 patients with GERD and with 19 healthyvolunteers who received either omeprazole 20 mgtwice daily or placebo. The mean intragastric pHwas found to be similar between the patients withpersistent symptoms who were receiving omeprazole20 mg twice daily and the healthy subjects receivingplacebo. Gastric pH monitoring in 7 patients given80 mg of omeprazole daily, however, demonstrated asignificant reduction in the mean percentage oftime that the pH was less than 4, indicating thatresponse is often a dose-dependent phenomenon.

Up to 70% of healthy subjects given twice-dailyPPI therapy experience an intragastric pH less than4 for more than 1 hour overnight (between 10:00PM and 6:00 AM). Katz and colleagues81 noted thatnocturnal acid breakthrough in patients who takeomeprazole 20 mg twice daily is often accompaniedby esophageal reflux and, therefore, esophageal acidexposure. Of 61 patients with GERD, 70% experi-enced nocturnal acid breakthrough and 33% expe-rienced nighttime esophageal acid exposure. Of 15patients with Barrett’s esophagus, 80% experiencednocturnal acid breakthrough and 50% experiencedesophageal acid exposure. In the control group(patients without GERD), these percentages were67% (not significantly different) and 8% (P < 0.03),respectively. Nocturnal acid breakthrough accom-

TABLE 5Profile of PPIs in the maintenance of erosive esophagitis healing

PPI FDA approval Prescribed Efficacydate dose (mg/d) (%)*

Omeprazole71 1989 10 5820 77

Lansoprazole72 1995 15 67–7930 55–90

Rabeprazole73 1999 10 7520 87

Pantoprazole74 2000 20 70–7240 83–86

Esomeprazole70,75 2001 20 9340 94

* In controlled clinical trials for maintenance of healing.



■ REFERENCES1. DeVault KR, Castell DO. Updated guidelines for the diagnosis

and treatment of gastroesophageal reflux disease. Am J Gastro-enterol 1999; 94:1434–1442.

2. Dent J, Brun J, Fendrick AM, et al. An evidence-based appraisalof reflux disease management: the Genval Workshop Report. Gut1999; 44(suppl 2):1–16.

3. Ramirez B, Richter JE. Review article: promotility drugs in thetreatment of gastro-oesophageal reflux disease. Aliment Phar-macol Ther 1993; 7:5–20.

4. Thanik KD, Chey WY, Shah AN, Gutierrez JG. Refluxesophagitis: effect of oral bethanechol on symptoms and endo-scopic findings. Ann Intern Med 1980; 93:805–808.

5. Paull A, Grant AK. A controlled trial of metoclopramide inreflux esophagitis. Med J Aust 1974; 2:627–629.

6. Venables CW, Bell D, Eccleston D. A double-blind study ofmetoclopramide in symptomatic peptic esophagitis. Postgrad MedJ 1973; 49(suppl 4):73–77.

7. McCallum RW, Ippoliti AF, Cooner C, et al. A controlled trialof metoclopramide in symptomatic gastro-esophageal reflux. NEngl J Med 1977; 296:354–357.

8. Bright-Asare P, El-Bassoussi M. Cimetidine, metoclopramide orplacebo in the treatment of symptomatic gastroesophageal reflux.J Clin Gastroenterol 1980; 2:149–156.

9. Temple JG, Bradby GVH, O’Connor F, et al. Cimetidine and met-oclopramide in esophageal reflux disease. BMJ 1983; 286:1863–1864.

10. Guslandi M, Testoni PA, Passaretti S, et al. Ranitidine vsmetoclopramide in the medical treatment of reflux esophagitis.Hepatogastroenterology 1983; 30:96–98.

11. Tytgat GNJ, Nio CY. The medical therapy of reflux oesophagi-tis. Baillieres Clin Gastroenterol 1987; 1:791–807.

12 . Blackwell JN, Heading RC, Fettes MR. Effects of domperidoneon lower oesophageal sphincter pressure and gastro-oesophagealreflux in patients with peptic esophagitis. Progress with domperi-done. Royal Society of Medicine. International Congress andSymposium Series 1981; 36:57–65.

13. Masci E, Testoni PA, Passaretti S, Guslandi M, Tittobello A.Comparison of ranitidine, domperidone maleate, and ranitidine +domperidone maleate in the short-term treatment of reflux esoph-

agitis. Drugs Exp Clin Res 1985; 11:687–692.14. Guslandi M, Dell’Oca M, Molteni V, et al. Famotidine vs dom-

peridone, versus a combination of both in the treatment of refluxesophagitis. Gastroenterology 1989; 96(suppl):191A. Abstract.

15. Richter JE. Cisapride: limited access and alternatives. Cleve ClinJ Med 2000; 67:471–472.

16. DeMicco M, Berenson M, Wu W, et al. Cisapride in the treatmentof GERD: a double-blind, placebo-controlled multicenter dose-response trial. Gastroenterology 1992; 102(suppl):59A. Abstract.

17. Galmiche JP, Fraitag B, Filoche B, et al. Double-blind compari-son of cisapride and cimetidine in treatment of reflux esophagitis.Dig Dis Sci 1990; 35:649–655.

18. Elsborg L, Beck B, Stubgaard M. Effect of sucralfate on gastro-esophageal reflux in esophagitis. Hepatogastroenterology 1985;32:181–184.

19. Katz PO, Castell DO. Gastroesophageal reflux disease duringpregnancy. Gastroenterol Clin North Am 1998; 27:153–167.

20. Simon B, Mueller P. Comparison of the effect of sucralfate andranitidine in reflux esophagitis. Am J Med 1987; 83:43–47.

21. Hameeteman W, van de Boomgaard DM, Dekker W, et al.Sucralfate versus cimetidine in reflux esophagitis: single-blindmulticenter study. J Clin Gastroenterol 1987; 9:390–394.

22. Laitinen S, Stahlberg M, Kairaluoma MI, et al. Sucralfate andalginate/antacid in reflux esophagitis. Scand J Gastroenterol 1985;20:229–232.

23. Chiba N, De Gara CJ, Wilkinson JM, Hunt RH. Speed of heal-ing and symptom relief in grade II to IV gastroesophageal refluxdisease: a meta-analysis. Gastroenterology 1997; 112:1798–1810.

24. Simon B, Ravelli GP, Goffin H. Sucralfate gel versus placebo inpatients with non-erosive gastro-oesophageal reflux disease.Aliment Pharmacol Ther 1996; 10:441–446.

25. Kitchin LI, Castell DO. Rationale and efficacy of conservativetherapy for gastroesophageal reflux disease. Arch Intern Med 1991;151:448–454.

26. Klinkenberg-Knol EC, Festen HPM, Meuwissen SGM. Phar-macological management of gastro-oesophageal reflux disease.Drugs 1995; 49:695–710.

27. Furman D, Mensh R, Winan G, et al. A double-blind trial com-paring high dose liquid antacid to placebo and cimetidine inimproving symptoms and objective parameters in gastroesophageal

panied by esophageal reflux in patients on twice-daily PPI therapy has important implications formedical therapy in patients with severe GERD orBarrett’s esophagus.

As previously discussed, Peghini and colleagues62

found that the combination of an H2RA and a PPImay be effective in managing nocturnal acid break-through. A recent study by Fackler and colleagues82

yielded differing results. The Fackler study showedthat, because tolerance developed to 300 mg of ran-itidine in patients with nocturnal acid break-through, there was no difference in acid suppressionbetween omeprazole 20 mg twice daily and omepra-zole 20 mg twice daily plus ranitidine after 1 week ofcombination therapy.

SafetyIn general, PPIs are a remarkably safe and well-tol-erated class of drugs. Headache and diarrhea are the

most frequently reported side effects, and theydevelop at rates that do not differ significantly fromthose in placebo-treated patients.83 Serious sideeffects resulting from PPI treatment are rare.Moreover, few clinically relevant drug interactionshave been reported. The PPIs’ benign safety profilehas contributed to their widespread prescriptionand use.

The long-term safety of PPIs has been a topic ofsome debate because most randomized controlledtrials of their effects have not extended beyond 1year. Thjodleifsson and colleagues84 addressed thisconcern in a 5-year controlled trial comparing effi-cacy and safety between rabeprazole and omepra-zole. They concluded from their results that bothPPIs were safe and well tolerated during the 5-yearstudy period, contributing to a growing body of evi-dence confirming the long-term safety of PPIs inacid suppression.


M AT O N

reflux. Gastroenterology 1992; 82(suppl):A1062. Abstract.28. Tougas G, Armstrong D. Efficacy of H2-receptor antagonists in

the treatment of gastroesophageal reflux disease and its symptoms.Can J Gastroenterol 1997; 11(suppl B):51–54.

29. Lin JH. Pharmacokinetic and pharmacodynamic properties of his-tamine H2-receptor antagonists: relationship between intrinsicpotency and effective plasma concentrations. Clin Pharmacokinet1991; 20:218–236.

30. Hallerback B, Glise H, Johansson B, et al. Gastro-oesophagealreflux symptoms: clinical findings and effect of ranitidine treat-ment. Eur J Surg Suppl 1998; 583:6–13.

31. Kahrilas PJ, Fennerty MB, Joelsson B. High- versus standard-dose ranitidine for control of heartburn in poorly responsive acidreflux disease: a prospective, controlled trial. Am J Gastroenterol1999; 94:92–97.

32. McCarty-Dawson D, Sue SO, Morrill B, Murdock RH. Raniti-dine versus cimetidine in the healing of erosive esophagitis. ClinTher 1997; 18:1150–1160.

33. Koelz HR, Bircher R, Bretholz A, et al. Healing and relapse ofreflux esophagitis during treatment with ranitidine. Gastroen-terology 1986; 91:1198–1205.

34. Wilder-Smith C, Halter F, Ernst T, et al. Loss of acid suppressionduring dosing with H2-receptor antagonists. Aliment PharmacolTher 1991; 4(suppl 1):15–27.

35. Scott M, Gelhot A. Gastroesophageal reflux disease: diagnosisand management. Am Fam Phys 1999; 59:1161–1169.

36. Colin-Jones DG. The role and limitations of H2-receptor antag-onists in the treatment of gastro-esophageal reflux disease. Ali-ment Pharmacol Ther 1995; 9(suppl 1):9–14.

37. Lau JYW, Sung JJY, Lee KKC, et al. Effect of intravenousomeprazole on recurrent bleeding after endoscopic treatment ofbleeding peptic ulcer. N Engl J Med 2000; 343:310–316.

38. Kahrilas PJ, Falk GW, Johnson DA, et al. Esomeprazoleimproves healing and symptom resolution as compared withomeprazole in reflux esophagitis patients: a randomized controlledtrial. Aliment Pharmacol Ther 2000; 14:1249–1258.

39. Richter JE, Kahrilas PJ, Johanson J, et al. Efficacy and safety ofesomeprazole compared with omeprazole in GERD patients witherosive esophagitis: a randomized controlled trial. Am JGastroenterol 2001; 96:656–665.

40. Lew EA. Review article: pharmacokinetic concerns in the selec-tion of anti-ulcer therapy. Aliment Pharmacol Ther 1999; 13(suppl5):11–16.

41. Maton PN. Omeprazole. N Engl J Med 1991; 324:965–975. 42. Fitton A, Wiseman L. Pantoprazole: a review of its pharmacolog-

ical properties and therapeutic use in acid-related disorders. Drugs1996; 51:460–482.

43. Physicians Desk Reference. 54th ed. Montvale, N.J.: MedicalEconomics Company; 2000.

44. Robinson M, Sloan S, Gardner J, Barth J. Gastric acid suppres-sion by rabeprazole: how much is enough to normalize exophagealacid exposure in gastroesophageal reflux disease? Gastroen-terology 1996; 116(suppl):A292–A293.

45. Hassan-Alin M, Nizai M, Röhss K, et al. Esomeprazole, the S-iso-mer of omeprazole, is optically stable in humans. Gastroenterology2000; 118(suppl):A1244–A1245. Abstract.

46. Hassan-Alin M, Röhss K, Andersson T, Nyman L. Pharmaco-kinetics of esomeprazole after oral and intravenous administrationof single and repeated doses to healthy subjects. Gastroenterology2000; 118(suppl):A16. Abstract.

47. Spencer CM, Faulds D. Esomeprazole. Drugs 2001; 60:321–329.48. Carlsson R, Galmiche J-P, Dent J, et al. Prognostic factors influ-

encing relapse of oesophagitis during maintenance therapy withantisecretory drugs: a meta-analysis of long-term omeprazole tri-als. Aliment Pharmacol Ther 1997; 11:473–482.

49. Venables TL, Newland RD, Patel AC, et al. Omeprazole 10 mil-ligrams once daily, omeprazole 20 milligrams once daily, or raniti-dine 150 milligrams twice daily, evaluated as initial therapy for

the relief of symptoms of gastro-oesophageal reflux disease. ScandJ Gastroenterol 1997; 32:965–973.

50. Richter JE, Campbell DR, Kahrilas PJ, et al. Lansoprazole com-pared with ranitidine for the treatment of nonerosive gastro-esophageal reflux disease. Arch Intern Med 2000; 160:1803–1809.

51. Castell DO, Richter JE, Robinson M, et al. Efficacy and safetyof lansoprazole in the treatment of erosive reflux esophagitis. AmJ Gastroenterol 1996; 91:1749–1757.

52. Mössner J, Hölscher AH, Herz R, Schneiders A.A double-blind studyof pantoprazole and omeprazole in the treatment of reflux oesophagi-tis: a multicentre trial. Aliment Pharmacol Ther 1995; 9:321–326.

53. Dekkers CPM, Beker JA, Thjodleifsson B, et al. Double-blind,placebo-controlled comparison of rabeprazole 20 mg vs. omeprazole20 mg in the treatment of erosive or ulcerative gastro-oesophagealreflux disease. Aliment Pharmacol Ther 1999; 13:49–57.

54. Richter JE, Kahrilas PJ, Hwang C, Marino V, Hamelin B. Eso-meprazole is superior to omeprazole for the healing of erosiveesophagitis in GERD patients. Gastroenterology 2000; 118(4 suppl2):A20. Abstract 343.

55. Lind T, Havelund T, Carlsson O, et al. Heartburn withoutoesophagitis: efficacy of omeprazole therapy and features determin-ing therapeutic response. Scand J Gastroenterol 1997; 32:974–979.

56. Bate CM, Griffin SM, Keeling PWN, et al. Reflux symptomrelief with omeprazole in patients without unequivocal oesophagi-tis. Aliment Pharmacol Ther 1996; 10:547–555.

57. Carlsson R, Dent J, Watts R, et al. Gastro-oesophageal reflux diseasein primary care: an international study of different treatment strate-gies with omeprazole. Eur J Gastroenterol Hepatol 1998; 10:119–124.

58. Lind T, Havelund T, Lundell L, et al. On-demand therapy withomeprazole for the long-term management of patients with heart-burn without oesophagitis—a placebo-controlled randomizedtrial. Aliment Pharmacol Ther 1999; 13:907–914.

59. Talley NJ, Lauritsen K, Tunturi-Hihnalas H, et al. Esomeprazole20 mg maintains symptom control in endoscopy-negative gastro-oesophageal reflux disease: a controlled trial of ‘on-demand’ thera-py for 6 months. Aliment Pharmacol Ther 2001; 15:347–354.

60. Tygat GNJ. Review article: treatment of mild and severe cases ofGERD. Aliment Pharmacol Ther 2002; 16(suppl 4):73–78.

61. Khoury RM, Katz PO, Hammod R, Castell DO. Bedtime rani-tidine does not eliminate the need for a second daily dose ofomeprazole to suppress nocturnal gastric pH. Aliment PharmacolTher 1999; 13:675–678.

62. Peghini PL, Katz PO, Castell DO. Ranitidine controls nocturnalgastric acid breakthrough on omeprazole: a controlled study innormal subjects. Gastroenterology 1998; 115:1335–1339.

63. Caro JJ, Salas M, Ward A. Healing and relapse rates in gastro-esophageal reflux disease treated with the newer proton-pumpinhibitors lansoprazole, rabeprazole, and pantoprazole comparedwith omeprazole, ranitidine, and placebo: evidence from random-ized clinical trials. Clin Ther 2001; 23:998–1017.

64. Lundell LR, Dent J, Bennett JR, et al. Endoscopic assessment ofoesophagitis: clinical and functional correlates and further valida-tion of the Los Angeles classification. Gut 1999; 45:172–180.

65. Sontag SJ, Hirschowitz BJ, Holt S, et al. Two doses of omepra-zole versus placebo in symptomatic erosive esophagitis: the USmulticenter study. Gastroenterology 1992; 102:109–118.

66. Edwards SJ, Lind T, Lundell L. Systematic review of protonpump inhibitors for the acute treatment of reflux oesophagitis.Aliment Pharmacol Ther 2001; 15:1729–1736.

67. Röhss K, Wilder-Smith C, Claar-Nilsson C, et al. Esomeprazole40 mg provides more effective acid control than standard doses ofall other proton pump inhibitors. Gastroenterology 2001;120(suppl):A419. Abstract.

68. Castell DO, Kahrilas MD, Richter JE, et al. Esomeprazole (40mg) compared with lansoprazole (30 mg) in the treatment of ero-sive esophagitis. Am J Gastroenterol 2002; 97:575–583.

69. Lauristen K, Devière J, Bigard M-A, et al. Esomeprazole 20 mg andlansosprazole 15 mg in maintaining healed reflux esophagitis:

Metropole study results. Aliment Pharmacol Ther 2003; 17:333–341.70. Johnson DA, Benjamin SB, Whipple J, D’Amico D, Hamelin B.

Esomeprazole once daily for 6 months is effective therapy for main-taining healed erosive esophagitis and for controlling gastroesophagealreflux disease: a randomized, double-blind, placebo-controlled study ofefficacy and safety. Am J Gastroenterol 2001; 96:27–34.

71. Prilosec [package insert]. Wilmington, Del.: AstraZeneca Phar-maceuticals LP; 2002.

72. Prevacid [package insert]. Lake Forest, Ill.: TAP PharmaceuticalProducts Inc.; 2002.

73. Aciphex [package insert]. Teaneck, N.J.: Eisai Inc.; 2002.74. Protonix [package insert]. Philadelphia: Wyeth Laboratories; 2003.75. Nexium [package insert]. Wilmington, Del.: AstraZeneca Phar-

maceuticals LP; 2003.76. Protonix IV [package insert]. Philadelphia: Wyeth Pharmaceuti-

cals Inc.; 2003. 77. DeVault KR, Castell DO. Guidelines for the diagnosis and treat-

ment of gastroesophageal reflux disease. Arch Intern Med 1995;155:2165–2173.

78. Freston JW. The management of gastroesophageal reflux diseasein the era of proton pump inhibitors. Pract Gastroenterol 1994;28(suppl):S24–S30.

79. Holloway RH, Dent J, Narielvala F, Mackinnon AM. Relationbetween oesophageal acid exposure and healing of oesophagitiswith omeprazole in patients with severe reflux oesophagitis. Gut1996; 38:649–654.

80. Leite LP, Johnston BT, Just RJ, Castell DO. Persistent acidsecretion during omeprazole therapy: a study of gastric acid pro-files in patients demonstrating failure of omeprazole therapy. AmJ Gastroenterol 1996; 91:1527–1531.

81. Katz PO, Anderson C, Khoury R, Castell DO. Gastro-oesophagealreflux associated with nocturnal gastric acid breakthrough on protonpump inhibitors. Aliment Pharmacol Ther 1998; 12:1231–1234.

82. Fackler WK, Ours TM, Vaezi MF, Richter JE. Long-term effectof H2RA therapy on nocturnal gastric acid breakthrough. Gastro-enterology 2002; 122:625–632.

83. Spechler SJ. In: Feldman M, Friedman LS, Sleisenger MH, eds.Sleisenger & Fordtran’s Gastrointestinal and Liver Disease. 7thed. Philadelphia: Saunders; 2002:747–781.

84. Thjodleifsson B, Rindi G, Fiocca A, et al. A randomized, dou-ble-blind trial of the efficacy and safety of 10 or 20 mg rabeprazolecompared with 20 mg omeprazole in the maintenance of gastro-oesophageal reflux disease over 5 years. Aliment Pharmacol Ther2003; 17:343–351.



cleveland clinic journal of medicine-2003-maton-s51

Documents