Advertisement

Gastric pH and Therapeutic Responses to Exsomeprazole in Patients With Functional Dyspepsia: Potential Clinical Implications

Published:October 14, 2016DOI:https://doi.org/10.1016/j.amjms.2016.09.010

      Abstract

      Background

      Therapy for functional dyspepsia remains a challenge. This study aimed to evaluate esomeprazole (E) versus placebo (P) regarding (1) the effectiveness in providing relief of abdominal pain or discomfort during 16 weeks of therapy in patients with functional dyspepsia having moderate or severe symptoms; (2) the effects on gastric acid suppression and (3) the relationship between symptom relief and gastric pH.

      Methods

      Enrolled patients were randomized to E (n = 38) or P (n = 35) in a double-blind, placebo-controlled trial. Outcomes were measured at four 4-week intervals. Drug dose titrated at each visit, based on relief of abdominal symptoms. The 24-hour gastric pH was monitored at baseline, 4 and 8 weeks.

      Results

      After 4 weeks, 71% of E patients (40 mg) reported satisfactory symptom relief versus 34% of P patients (P < 0.001). When the dose for nonresponders (NR) was titrated to 40 mg twice daily, the E relief rate increased to 82% versus 56% in P group (P < 0.05). During the next 4 weeks, with dose decreased by half in responders, E response rate declined to 69% versus 48% in P group (P < 0.10). When the dose was increased for NR during the last 4 weeks, E rate increased to 83% versus 57% in P group (P < 0.05). At 4 and 8 weeks for E responders and NR, patients׳ pH >4 value increased significantly compared to baseline.

      Conclusions

      (1) Though E 40 mg once daily is superior to P, some patients benefit from 40 mg twice daily; (2) E, 40 mg once daily, profoundly inhibits gastric acid secretion; (3) intragastric pH monitoring before and after therapy may help address the relationship between symptomatic relief and gastric acid secretion and (4) some patients respond to monitored titrated placebo therapy.

      Key Indexing Terms

      Introduction

      Dyspepsia is a term used to describe pain or discomfort in the upper abdomen unrelated to defecation or stool frequency. The U.S. prevalence rate reaches 2.5%.
      • Brun R.
      • Kuo B.
      Functional dyspepsia.
      • Lacy B.E.
      • Talley N.J.
      • Locke 3rd, G.R.
      • et al.
      Review article: current treatment options and management of functional dyspepsia.
      The pathophysiology of dyspepsia is unclear. Theories include Helicobacter pylori infection, medications (especially nonsteroidal anti-inflammatory drugs [NSAIDs]), gastric dysmotility, poor mucosal defense and excessive mucosal acid exposure.
      • Robinson M.
      Dyspepsia: challenges in diagnosis and selection of treatment.
      Rome II Criteria (Table 1) aids in the diagnosis of functional dyspepsia (FD).
      • Tack J.
      • Talley N.J.
      • Camilleri M.
      • et al.
      Functional gastroduodenal disorders.
      • Talley N.J.
      • Vakil N.B.
      • Moayyedi P.
      American Gastroenterological Association technical review on the evaluation of dyspepsia.
      Most patients with dyspepsia will have a normal physical examination, except for epigastric tenderness on palpation. A FD diagnosis requires that symptoms be present for at least 12 weeks in the preceding 12 months.
      • Lacy B.E.
      Functional dyspepsia and gastroparesis: one disease or two?.
      Approximately 75% of patients have FD with no identifiable underlying cause.
      • van Zanten S.V.
      • Armstrong D.
      • Chiba N.
      • et al.
      Esomeprazole 40 mg once a day in patients with functional dyspepsia: the randomized, placebo-controlled “ENTER” trial.
      • Thumshirn M.
      Pathophysiology of functional dyspepsia.
      Uninvestigated dyspepsia refers to patients with either new or recurrent dyspepsia symptoms that have not had any diagnostic investigations. Investigated dyspepsia is divided into organic dyspepsia or FD. Organic dyspepsia has a pathophysiological or anatomic explanation for the dyspepsia.
      • Brun R.
      • Kuo B.
      Functional dyspepsia.
      In contrast, FD is diagnosed after various investigations have been performed and found to be normal. FD does not include patients with a dominant compliant of heartburn or regurgitation.
      TABLE 1Rome II Criteria for functional dyspepsia.
      At least 12 weeks that does not have to be consecutive within the previous 12 months of:
      1. Persistent or recurrent symptoms of upper abdomen pain or discomfort.
      2. No evidence of organic disease that could possibly explain the symptoms.
      3. No evidence that the dyspepsia is relieved by defection or associated with a change in stool frequency or stool shape (rule out irritable bowel syndrome).
      When the local prevalence of FD is more than 10%, the initial approach to patients with dyspepsia involves screening for H pylori.
      • Moayyedi P.
      • Soo S.
      • Deeks J.
      • et al.
      Eradication of Helicobacter pylori for non–ulcer dyspepsia. Cochrane.
      The American Gastroenterological Association suggests that patients with dyspepsia must undergo screening for H pylori and eradication. Standard therapy is that patients with dyspepsia along with H pylori or after eradication should be started on proton pump inhibitors (PPIs).
      PPIs are very effective for the suppression of gastric acid secretion through the inhibition of the H+, K+ ATPase at the secretory surface of the parietal cells.
      • Vachhani R.
      • Olds G.
      • Velanovich V.
      Esomeprazole: a proton pump inhibitor.
      • Niu X.P.
      • Yu B.P.
      • Wang Y.D.
      • et al.
      Risk factors for proton pump inhibitor refractoriness in Chinese patients with non-erosive reflux disease.
      Esomeprazole (E), the S-isomer of omeprazole, is rapidly absorbed and gradually metabolized by cytochrome P450 (CYP) isoenzymes, particularly CYP2C19 and CYPC3A4.
      • McKeage K.
      • Blick S.K.
      • Croxtall J.D.
      • et al.
      Esomeprazole: a review of its use in the management of gastric acid-related diseases in adults.
      This results in a profound and long inhibition of gastric acid secretion during a 24-hour period.
      • Yang H.
      • Li J.
      • Zhao Q.
      • et al.
      Pharmacokinetics, pharmacodynamics, and safety of esomeprazole injection/infusion in healthy Chinese volunteers: a five-way crossover study.
      E is used for the treatment of gastroesophageal reflux disease (GERD), peptic ulcers and the eradication of H pylori infection as an addition to antimicrobial agents.
      • Rotman S.R.
      • Bishop T.F.
      Proton pump inhibitor use in the U.S. ambulatory setting, 2002-2009.
      PPIs have been shown to provide symptom relief in some, but not all, patients with dyspepsia or FD.
      • Reimer C.
      • Søndergaard B.
      • Hilsted L.
      • et al.
      Proton-pump inhibitor therapy induces acid-related symptoms in healthy volunteers after withdrawal of therapy.
      • Bytzer P.
      • van Zanten S.V.
      • Mattsson H.
      • et al.
      Partial symptom-response to proton pump inhibitors in patients with non-erosive reflux disease or reflux oesophagitis—a post hoc analysis of 5796 patients.
      The relationship between FD symptom relief and gastric acidity remains to be explored further. The objectives of the current study were to evaluate the efficacy of E 40 mg once daily (QD) or twice daily (BID) in achieving symptom relief in patients with FD with non–ulcer dyspepsia in a 16-week, placebo-controlled study and to explore the relationship of symptom relief with the degree of acid suppression.
      The current study was conducted with the following 3 specific aims:
      • (1)
        Evaluate the efficacy of E versus placebo (P) in providing satisfactory relief of pain or discomfort symptoms in patients with FD during 16 weeks of therapy.
      • (2)
        Evaluate the effect of E versus P on gastric acid during 24-hour pH monitoring.
      • (3)
        Assess the relationship between relief of dyspepsia symptoms and changes in gastric pH.

      Methods and Statistical Considerations

      Study Population

      The study was conducted in 73 H pylori–negative subjects (32 males, 41 females; 18-55 years old) with epigastric pain or discomfort related to non–ulcer dyspepsia and without alarm features. This study was conducted as a randomized, double-blind, placebo-controlled trial. Rome II Diagnostic Criteria for FD (Table 1) were used to determine patient eligibility. After the Institutional Review Board approved the protocol, all patients were presented with informed consent that had to be signed before subjects entered the clinical trial. An additional informed consent for the esophagogastroduodenoscopy was obtained before the procedure, if required.

      Inclusion Criteria

      • (1)
        18-55 years old.
      • (2)
        Diagnosis of epigastric pain or discomfort currently rated by the patient as moderate to severe in intensity (4-point scale), unrelated to exercise and present at least 3 times per week for 12 weeks.
      • (3)
        Patients may have other symptoms, including heartburn, regurgitation, bloating (abdominal distention), early satiety (feeling of fullness), belching (burping) or nausea. The dominant symptom must be epigastric pain or discomfort.
      • (4)
        Must be capable of and willing to give informed consent and comply with all study requirements.

      Exclusion Criteria

      • (1)
        H pylori–positive serology or had received eradication treatment within the last 12 months.
      • (2)
        Regular use of NSAIDs, acetylsalicylic acid or cyclooxygenase-2 selective inhibitors.
      • (3)
        Treatment with any medications such as antibiotics and codeine.
      • (4)
        History or presence of endoscopic or radiologic evidence of esophagitis, Barrett׳s esophagus, gastric ulcer or duodenal ulcer.
      • (5)
        History or presence of chronic gastritis, duodenal erosions or ulcers.
      • (6)
        History of previous gastrointestinal (GI) surgery.
      • (7)
        Presence of concomitant symptoms of irritable bowel syndrome assessed by 3 or more of the Manning or Rome criteria.
      • (8)
        History or presence of other known other diseases that might explain dyspepsia symptoms (e.g., biliary colic, hiatal hernia and peptic ulcer disease).
      • (9)
        Pregnancy or lactation.
      • (10)
        Regular consumption of greater than 1 alcoholic beverage per day.
      • (11)
        History of illicit substance abuse.
      • (12)
        Unwillingness or expected inability to tolerate the absence of antisecretory medications (antacids, H2-receptor antagonists, PPIs or other GI pharmacotherapy) for the period of the study.
      • (13)
        Previous history of GI malignancy, peptic ulcer disease, previous upper GI surgery or esophageal motility disorders.

      Screening

      A screening evaluation was performed up to 14 days before the baseline evaluation. It included:
      • (1)
        Written informed consent,
      • (2)
        Complete medical history and physical examination,
      • (3)
        Pregnancy test (urine) and
      • (4)
        H pylori serology test.
      Three or more of the additional following symptoms were required for enrollment: upper abdominal pain or discomfort before meals or when hungry, upper abdominal pain or discomfort at night (wakes from sleep), upper abdominal pain or discomfort relieved by food, upper abdominal pain or discomfort relieved by antacids, periodic upper abdominal pain or discomfort and well-localized upper abdominal pain or discomfort.

      Stratified Randomization

      Eligible patients with moderate or severe epigastric pain or discomfort were enrolled. Before the stratified randomization, all patients received P (1 capsule daily) for 4 weeks. Stratified randomization codes were assigned sequentially to enrolled patients. Patient and clinician blinding was maintained using identical-appearing P and E tablets (blister packs).

      Outcome Measures

      Patient assessment of dyspepsia symptom severity was determined at the beginning of the study using the validated 7-point Global Overall Symptom scale for dyspepsia. This consists of (1) no problem, (2) minimal problem (can be easily ignored), (3) mild problem (can be ignored with effort), (4) moderate problem (cannot be ignored, does not affect daily activities), (5) moderately severe problem (cannot be ignored, occasionally affects daily activities), (6) severe problem (cannot be ignored, affects daily activities) and (7) very severe problem. The Quality of Life in Reflux and Dyspepsia (QOLRAD) questionnaire was also used at baseline, 4, 8, 12 and 16 weeks after starting the treatment. The QOLRAD questionnaire contains 25 itemized questions in 5 relevant domains—emotional distress (6 items), sleep disturbance (5 items), drink or food problems (6 items), social or physical functioning (5 items) and vitality (3 items). Each question has 7 options and results are reported as average scores for each domain. A higher score correlates with a better quality of life.
      Patients were instructed to rate the severity of dyspepsia (epigastric pain or discomfort) using validated 7-graded diary cards over the last 7 consecutive days of the 2 week run-in period and at the end of each treatment period (4, 8, 12 and 16 weeks). The efficacy of therapy was based on the symptom scores recorded on these diary cards. The diary cards ask “Please state for each day if you have experienced pain or discomfort in the stomach.” The global assessment of overall severity of symptoms was measured as secondary end points using a Likert 4 categorical scale (1 = symptom free, 2 = improved, 3 = unchanged and 4 = worse) at the end of each treatment period (4, 8, 12 and 16 weeks). A 7-point Likert scale was used to assess how much or often the item described the feeling of patient; degree of distress (none at all, hardly any at all, a little, some, a moderate amount, a lot and a great deal) and the frequency of the problem (none of the time, hardly any of the time, a little of the time, quite a lot of the time, most of the time and all of the time). A mean item score was calculated for each patient.
      This report presents information on responders (R) to randomized therapy versus nonresponders (NR). Patients were considered an R to the acid suppression therapy if they reported that their symptoms were no more than mild and these symptoms occurred on frequency of less than 2 days per week. An R to therapy was defined (mathematically) as a patient whose sum of symptom scores was 0 or 1. Outcome measures included relief of pain or discomfort symptoms, measured at 4 consecutive 4-week intervals (16 weeks total), a 7-day daily diary preceding each visit, a quality-of-life questionnaire and a global overall symptoms assessment. At each visit, patients were asked to rate any adverse effect that occurred as mild, moderate or severe.

      Randomization

      Patients (mean age of 44.2; 56% females) were randomized to P (n = 35) or E regimens (n = 38) 40 mg (Table 1). Investigators, study personnel and patients were blinded to therapy assignment until the study׳s completion. E and P tablets were identical in their appearance. Patients were initially randomized to E 40 mg or P 40 mg, and were instructed to take the study drug at least 1 hour before breakfast. Patients who required twice-a-day dosing of E 40 mg later in the course of the study were given 1 tablet in the morning and 1 in the evening. The study lasted a total of 16 weeks and the patient adjusted the dose of either active drug or P at 4-week intervals, based on self-reported symptom relief. Medication compliance was assessed by tracking the pill count of the returned medication at each visit.
      At baseline, before randomization, gastric acidity was evaluated by 24-hour pH monitoring using 2 electrodes placed at 5 and 10 cm below the lower esophageal sphincter. After 4 weeks of therapy, a second symptom score and global assessment of symptoms were measured using diary cards over the fourth weeks on treatment. All patients underwent a second 24-hour gastric pH-monitoring test after 4 weeks of therapy. Patients who failed to achieve satisfactory symptom relief at 4 weeks were continued on a double dose of P or E (40 mg twice a day), for an additional 4 weeks (8 weeks total). Patients with satisfactory symptom relief at 4 weeks were continued on the single daily dose of medication (E or P, blinded) for an additional 4 weeks (8 weeks). The pH monitoring was repeated after 8 weeks of therapy. Patients, who failed to achieve satisfactory symptom relief at 8 weeks, were continued on a double dose of either P or E (40 mg twice daily), for an additional 4 weeks (12 weeks total). Patients who achieved symptomatic relief at 8 weeks underwent a step-down dose adjustment (from 80-40 mg daily or from 40-20 mg daily) and continued on this dosage for another 4 weeks of therapy (12 weeks total). At 12 and 16 weeks, only outcome measures were reassessed. Patients who failed to achieve symptomatic relief at 12 weeks were continued on therapy for another 4 weeks (16 weeks total) with a double dose of either P or E (40 mg twice daily). Patients who achieved symptomatic relief underwent a step-down dose adjustment (from 80-40 mg daily or from 40-20 mg daily) and continued on this dosage for another 4 weeks of therapy (16 weeks total). If symptoms increased after step-down dose adjustment, the dose was increased to the dose that was effective in symptom relief during the last 4 weeks of therapy. Outcome measures were repeated after 16 weeks of therapy.
      Gastric pH was monitored using dual-probe catheter and pH Monitoring System (Sandhill Scientific, Milwaukee, WI). In the 24-hour pH monitoring recordings, daytime and nighttime diurnal median gastric pH were calculated. In addition, the percentage of time gastric pH remained >4.0 was evaluated and correlated with the degree of symptom relief and improvement of QOLRAD. Each patient had a binder with diaries, QOLRAD questionnaires and relevant forms for results of all conducted tests throughout the study protocol.

      Statistical Analysis

      The 95% CI or P values were calculated for all data analyses. A P < 0.05 was considered statistically significant on applied statistical tests. A graded 7-point Likert scale and the QOLRAD questionnaire was used at baseline, 4, 8, 12 and 16 weeks of therapy to assess whether the patient was an R versus NR. An R to therapy was defined as a patient whose sum of symptom scores was 0 or 1 at the end of each treatment interval. Descriptive statistics were used in the analysis of baseline characteristics. Mean, standard deviation and standard deviation from 25-75% values were obtained based on the time (minutes) that the gastric pH was >4 within the 24-hour period. The 24-hour pH monitoring was performed at baseline (Visit 2—V2), week 4 (Visit 3—V3) and week 8 (Visit 4—V4). Comparative data analyses were performed using t test, paired t test, Mann-Whitney rank-sum test and Wilcoxon signed-rank test as stated in Tables 2-5.

      Results

      The mean age of the patient population was 44.2 years, 56% of study subjects were female. After 4 weeks of therapy, 71% patients receiving E 40 mg capsule QD reported satisfactory relief of symptoms versus 34% on P (P < 0.001) (Figure 1, Visit 3). The rate of satisfactory relief of symptoms increased to 82% when visit 3 E NRs doubled their dose to 40 mg BID. The P response rate also increased to 56% (P < 0.05) on BID dosing among NR at visit 3 (Figure 1, Visit 4). During the next 4 weeks of therapy (dose in Rat visit 4) reduced to 20 or 40 mg QD, correspondingly), the E response rate declined to 69%, compared to a P response rate of 48% (P < 0.10) (Figure 1, Visit 5). When the dose at visit 5 was increased again for NR during the last 4 weeks of therapy, the response rate in patients on E increased to 83% versus 57% on P (P < 0.05) (Figure 1, Visit 6).
      FIGURE 1
      FIGURE 1Satisfactory relief of symptoms over the past 4 weeks.
      In the E patients, there was a statistical significant increase of pH > 4 compared to baseline (V2) at both the proximal and distal channels at 4 weeks (V3; second pH test) and at 8 weeks (V4; third pH test; P <0.001; Table 2).
      TABLE 2Results of pH > 4 values based on 24-hour gastric pH monitoring before (V2) and after 4 weeks (V3) and 8 weeks (V4) of therapy with esomeprazole or placebo in patients with non–ulcer dyspepsia.
      pH > 4 Channel
      ProximalDistal
      Esomeprazole
      V2 (baseline)—first pH test (n = 38)
       Mean (SEM)637 (69.1)194 (32.1)
       95% CI(497, 777)(129, 259)
      V3—second pH test (n = 38)
       Mean (SEM)1,155 (45.3)884 (59.9)
       95% CI(1,063, 1247)(763, 1,005)
      PP ≤ 0.001
      Wilcoxon signed-rank test.
      vs. esomeprazole baseline (first pH test)
      P ≤ 0.001
      Wilcoxon signed-rank test.
      vs. esomeprazole baseline (first pH test)
      V4—third pH test (n = 38)
       Mean (SEM)1,036 (67.3)796 (69.6)
       95% CI(900, 1,172)(655, 937)
      PP ≤ 0.001
      Paired t test.
      vs. esomeprazole baseline (first pH test)
      P ≤ 0.001
      Wilcoxon signed-rank test.
      vs. esomeprazole baseline (first pH test)
      P = 0.03
      Wilcoxon signed-rank test.
      vs. esomeprazole V3 (second pH test)
      P = 0.004
      Wilcoxon signed-rank test.
      vs. esomeprazole V3 (second pH test)
      Placebo
      V2 (baseline)—first pH test (n = 35)
       Mean (SEM)820 (72.3)252 (49.2)
       95% CI(673, 967)(152, 352)
      V3—second pH test (n = 35)
       Mean (SEM)794 (70.5)243 (39.4)
       95% CI(651, 937)(163, 323)
      PP = 0.44
      Wilcoxon signed-rank test.
      vs. placebo baseline (first pH test)
      P = 0.99
      Wilcoxon signed-rank test.
      vs. placebo baseline (first pH test)
      V4—third pH test (n = 35)
       Mean (SEM)764 (73.9)278 (61.4)
       95% CI(614, 914)(153, 403)
      PP = 0.50
      Paired t test.
      vs. placebo baseline (first pH test)
      P = 0.81
      Wilcoxon signed-rank test.
      vs. placebo baseline (first pH test)
      P = 0.66
      Paired t test.
      vs. placebo V3 (second pH test)
      P = 0.31
      Wilcoxon signed-rank test.
      vs. placebo V3 (second pH test)
      SEM, standard error of mean; V2, visit 2; V3, visit 3; V4, visit 4.
      a Paired t test.
      b Wilcoxon signed-rank test.
      In the P patients, there was no statistically significant increase of pH > 4 compared to baseline at either the proximal or distal channel at 4 weeks (V3; second pH test) or 8 weeks (V4; third pH test) (Table 2).
      The baseline mean values of gastric pH monitoring (V2) in patients randomized to E or P were similar (Table 3). After 4 weeks of either E or P, our data demonstrate a statistically significant difference in (P < 0.001) in pH > 4 values between E and P at both the proximal and distal channels at 4 weeks of therapy (V3; second pH test) and 8 weeks of therapy (V4; third pH test) (Table 3). Comparing E versus P patients at the proximal channel at 4 weeks (V3; second pH test), the increased difference in time of pH > 4 value was 46% and at 8 weeks it was 36%. Comparing the E versus P patients at the distal channel at 4 weeks of therapy (V3; second pH test), the increased difference in time of pH > 4 value was 264% and at 8 weeks of therapy (V4; third pH test), it was 186% (Table 3).
      TABLE 3Results of pH > 4 values based on 24-hour gastric pH monitoring before (V2) and after 4 weeks (V3) and 8 weeks (V4) of therapy with esomeprazole vs. placebo in patients with non–ulcer dyspepsia.
      pH > 4 Channel
      ProximalDistal
      V2 (Baseline) pH test
      Placebo (n = 35)
       Mean (SEM)820 (72.3)252 (49.2)
       95% CI(673, 967)(152, 352)
      Esomeprazole (n = 38)
       Mean (SEM)637 (69.1)194 (32.1)
       95% CI(497, 777)(129, 259)
      PP = 0.054
      Mann-Whitney rank-sum test.
      vs. placebo baseline
      P = 0.44
      Mann-Whitney rank-sum test.
      vs. placebo baseline
      V3 pH test
      Placebo (n = 35)
       Mean (SEM)794 (70.5)243 (39.4)
       95% CI(651, 937)(163, 323)
      PP = 0.44
      Wilcoxon signed-rank test.
      vs. placebo baseline
      P = 0.99
      Wilcoxon signed-rank test.
      vs. placebo baseline
      Esomeprazole (n = 38)
       Mean (SEM)1,155 (45.3)884 (59.9)
       95% CI(1,063, 1,237)(763, 1,005)
      PP ≤ 0.001
      Mann-Whitney rank-sum test.
      vs. placebo V3
      P ≤ 0.001
      Mann-Whitney rank-sum test.
      vs. placebo V3
      P ≤ 0.001
      Wilcoxon signed-rank test.
      vs. esomeprazole baseline
      P ≤ 0.001
      Wilcoxon signed-rank test.
      vs. esomeprazole baseline
      V4 pH test
      Placebo (n = 35)
       Mean (SEM)764 (73.9)279 (61.4)
       95% CI(614, 904)(154, 404)
      PP = 0.50
      Paired t test.
      vs. placebo baseline
      P = 0.81
      Wilcoxon signed-rank test.
      vs. placebo baseline
      P = 0.66
      Paired t test.
      vs. placebo V3
      P = 0.31
      Wilcoxon signed-rank test.
      vs. placebo V3
      Esomeprazole (n = 38)
       Mean (SEM)1,036 (67.3)796 (69.6)
       95% CI(900, 1,172)(655, 937)
      PP = 0.01
      Mann-Whitney rank-sum test.
      vs. placebo V4
      P ≤ 0.001
      Mann-Whitney rank-sum test.
      vs. placebo V4
      P ≤ 0.001
      Paired t test.
      vs. esomeprazole baseline
      P ≤ 0.001
      Wilcoxon signed-rank test.
      vs. esomeprazole baseline
      SEM, standard error of mean; V2, visit 2; V3, visit 3; V4, visit 4.
      a Paired t test.
      b Wilcoxon signed-rank test.
      c Mann-Whitney rank-sum test.
      We also considered whether R and NR differed in pH > 4 changes in response to therapy. There was no statistical significance in the pH > 4 value at the proximal or distal channel in subjects who were either an R or NR to E therapy at baseline (Table 4 and Figure 2). As noted, the time pH > 4 was significantly increased in E patients at subsequent visits. This increase did not appear to differ in E R and NR.
      TABLE 4Results of pH > 4 values based on 24-hour gastric pH monitoring before (V2) and after 4 weeks (V3) and 8 weeks (V4) of therapy with esomeprazole in nonresponders vs. responders in patients with non–ulcer dyspepsia.
      pH > 4
      Channel
      ProximalDistal
      Esomeprazole
      Baseline pH test
      Nonresponders (n = 11)
       Mean (SEM)628 (155)177 (37.7)
       95% CI(282, 974)(93, 261)
      Responders (n = 27)
       Mean (SEM)641 (76.2)201 (42.7)
       95% CI(484, 798)(114, 289)
      PP = 0.52
      Mann-Whitney rank-sum test.
      vs. V3-baseline nonresponders
      P = 0.85
      Mann-Whitney rank-sum test.
      vs. V3-baseline nonresponders
      V3 pH test
      Nonresponders (n = 11) (QD—all patients)
       Mean (SEM)1,121 (95)850 (141)
       95% CI(929, 1,333)(537, 1,163)
      PP = 0.02
      Wilcoxon signed-rank test.
      vs. V3-baseline nonresponders
      P = 0.001
      Paired t test.
      vs. V3-baseline nonresponders
      Responders (n = 27) (QD—all patients)
       Mean (SEM)1,170 (51.6)898 (63.5)
       95% CI(1,064, 1276)(767, 1,029)
      PP = 0.65
      Mann-Whitney rank-sum test.
      vs. V3 pH test nonresponders
      P = 0.88
      Mann-Whitney rank-sum test.
      vs. V3 pH test nonresponders
      P ≤ 0.001
      Paired t test.
      vs. V3-baseline responders
      P ≤ 0.001
      Paired t test.
      vs. V3-baseline responders
      Baseline pH test
      Nonresponders (n = 8)
       Mean (SEM)663 (176)239 (121)
       95% CI(246, 1,080)(10, 526)
      PP = 0.97
      Mann-Whitney rank-sum test.
      vs. V3-baseline nonresponders
      P = 0.77
      Mann-Whitney rank-sum test.
      vs. V3-baseline nonresponders
      Responders (n = 30)
       Mean (SEM)630 (75.6)182 (45.3)
       95% CI(475, 785)(89, 275)
      PP = 0.86
      Mann-Whitney rank-sum test.
      vs. V3-baseline responders
      P = 0.84
      Mann-Whitney rank-sum test.
      vs. V3-baseline responders
      P = 0.73
      Mann-Whitney rank-sum test.
      vs. V4-baseline nonresponders
      P = 0.71
      t Test.
      vs. V4-baseline nonresponders
      V4 pH test
      Nonresponders (n = 8) (BID—all patients)
       Mean (SEM)976 (168.6)614 (168.3)
       95% CI(577, 1,374)(216, 1,012)
      PP = 0.22
      Wilcoxon signed-rank test.
      vs. V4-baseline nonresponders
      P = 0.12
      Paired t test.
      vs. V4-baseline nonresponders
      P = 0.65
      Mann-Whitney rank-sum test.
      vs. V3 pH test nonresponders
      P = 0.24
      Mann-Whitney rank-sum test.
      vs. V3 pH test nonresponders
      Responders (n = 30) (QD—all patients)
       Mean (SEM)1,050 (74.3)839 (75.8)
       95% CI(898, 1,202)(684, 994)
      PP = 0.66
      Mann-Whitney rank-sum test.
      vs. V4 pH test nonresponders
      P = 0.22
      t Test.
      vs. V4 pH test nonresponders
      P ≤ 0.001
      Paired t test.
      vs. V4-baseline responders
      P ≤ 0.001
      Wilcoxon signed-rank test.
      vs. V4-baseline responders
      P = 0.5
      Mann-Whitney rank-sum test.
      vs. V3 pH test responders
      P = 0.70
      Mann-Whitney rank-sum test.
      vs. V3 pH test responders
      BID, twice daily; QD, once daily; SEM, stanadard error of mean; V2, visit 2; V3, visit 3; V4, visit 4.
      a Mann-Whitney rank-sum test.
      b Wilcoxon signed-rank test.
      c Paired t test.
      d t Test.
      FIGURE 2
      FIGURE 2A, Results of pH > 4 values at the proximal channel in subjects who were either responders (R) or nonresponders (NR) on esomeprazole or placebo therapy. B, Results of pH > 4 values at the distal channel in subjects who were either responders (R) or nonresponders (NR) on esomeprazole or placebo therapy.
      Subjects that received E and were R at either 4 weeks (V3; second pH test) or 8 weeks (V4; third pH test) had a statistical significance (P < 0.001) in pH > 4 value at the proximal and distal channels in comparison with their baseline. The subjects that were NR also demonstrated an increase in pH > 4 time value, but this was less significant at the proximal channel (P = 0.02; Table 4).
      At the fourth week (V3; second pH test) of E therapy, the R had an increased duration of pH > 4 value at the proximal channel by 82% and at the distal channel of 34% compared to the baseline time pH value. On analysis of the eighth week (V4; third pH test) of E therapy, the R had an increased duration of pH > 4 value at the proximal channel by 67% and at the distal channel of 361% compared to the baseline time pH value. During the eighth week of E therapy, the difference in time pH > 4 value between the baseline and V4 third pH test of NR was not statistically significant (P = 0.22) at proximal channel and also nonsignificant to distal channel (P = 0.12). This demonstrates a relationship between the acid suppression by E and symptomatic response to therapy. There was no statistical significance in the pH > 4 value at the proximal or distal channel in either R or NR to P therapy (Table 5 and Figure 2B). The only exception occurred in the comparison of R to P at the fourth week (V3; second pH test) at the distal channel with pH > 4 values increased by 47% compared to baseline values (P = 0.03).
      TABLE 5Results of pH > 4 values based on 24-hour gastric pH monitoring before (V2) and after 4 weeks (V3) and 8 weeks (V4) of therapy with placebo in nonresponders vs. responders in patients with non–ulcer dyspepsia.
      PlacebopH > 4
      Channel
      ProximalDistal
      V3 (baseline) pH test
      Nonresponders (n = 23)
       Mean (SEM)850 (85.1)284 (69.4)
       95% CI(674, 1,026)(140, 428)
      Responders (n = 12)
       Mean (SEM)763 (137.7)191 (52.3)
       95% CI(460, 1,066)(76, 306)
      PP = 0.74
      Mann-Whitney rank-sum test.
      vs. V3-baseline nonresponders
      P = 0.41
      Mann-Whitney rank-sum test.
      vs. V3-baseline nonresponders
      V3 pH test
      Nonresponders (n = 23) (QD—all patients)
       Mean (SEM)736 (80.5)183 (29.4)
       95% CI(569, 903)(122, 204)
      PP = 0.11
      Paired t test.
      vs. V3-baseline nonresponders
      P = 0.09
      Wilcoxon signed-rank test.
      vs. V3-baseline nonresponders
      Responders (n = 12) (QD—all patients)
       Mean (SEM)905 (134.8)360 (93.5)
       95% CI(608, 1,202)(154, 566)
      PP = 0.26
      t Test.
      vs. V3 pH test nonresponders
      P = 0.04
      Mann-Whitney rank-sum test.
      vs. V3 pH test nonresponders
      P = 0.57
      Wilcoxon signed-rank test.
      vs. V3-baseline responders
      P = 0.03
      Wilcoxon signed-rank test.
      vs. V3-baseline responders
      V4 (baseline) pH test
      Nonresponders (n = 16)
       Mean (SEM)872 (95.3)193 (39.3)
       95% CI(669, 1,075)(109, 277)
      PP = 0.98
      Mann-Whitney rank-sum test.
      vs. V3-baseline nonresponders
      P = 0.53
      Mann-Whitney rank-sum test.
      vs. V3-baseline nonresponders
      Responders (n = 19)
       Mean (SEM)777 (107.8)302 (83.7)
       95% CI(550, 1,004)(126, 478)
      PP = 0.87
      Mann-Whitney rank-sum test.
      vs. V3-baseline responders
      P = 0.47
      Mann-Whitney rank-sum test.
      vs. V3-baseline responders
      P = 0.52
      t Test.
      vs. V4-baseline nonresponders
      P = 0.59
      Mann-Whitney rank-sum test.
      vs. V4-baseline nonresponders
      V4 pH test
      Nonresponders (n = 16) (BID—all patients)
       Mean (SEM)766 (105)275 (84.5)
       95% CI(542, 990)(94.9, 455)
      PP = 0.21
      Paired t test.
      vs. V4-baseline nonresponders
      P = 0.85
      Wilcoxon signed-rank test.
      vs. V4-baseline nonresponders
      P = 0.83
      t Test.
      vs. V3 pH test nonresponders
      P = 0.51
      Mann-Whitney rank-sum test.
      vs. V3 pH test nonresponders
      Responders (n = 19) (QD—all patients)
       Mean (SEM)762 (101.8)282 (73.3)
       95% CI(462, 1,063)(90.0, 474)
      PP = 0.86
      Mann-Whitney rank-sum test.
      vs. V4 pH test nonresponders
      P = 0.68
      Mann-Whitney rank-sum test.
      vs. V4 pH test nonresponders
      P = 0.89
      Wilcoxon signed-rank test.
      vs. V4-baseline responders
      P = 0.57
      Wilcoxon signed-rank test.
      vs. V4-baseline responders
      P = 0.48
      Mann-Whitney rank-sum test.
      vs. V3 pH test responders
      P = 0.09
      Mann-Whitney rank-sum test.
      vs. V3 pH test responders
      BID, twice daily; QD, once daily; SEM, standard error of mean; V2, visit 2; V3, visit 3; V4, visit 4.
      a Mann-Whitney rank-sum test.
      b Paired t test.
      c Wilcoxon signed-rank test.
      d t Test.

      Discussion

      Our subjects were considered to be diagnosed with investigated FD. Screening determined that they were H pylori–negative, did not use NSAIDs and had a negative history of peptic ulcer disease with no anemia or episodes of upper gastrointestinal bleeding. The pathophysiology of FD is not well identified. However, multiple confounding factors are likely involved, such as physiologic, genetic, environmental and psychological factors. Pharmacologic management of function dyspepsia is difficult. The effectiveness of PPIs may vary depending on the pharmacokinetics and pharmacodynamics of the drug and the metabolism of the patient.
      • Brun R.
      • Kuo B.
      Functional dyspepsia.
      Current medications are aimed at only 1 target; however, multiple pathophysiological mechanisms may be involved in the effectiveness of acid suppression therapy. Therefore, understanding the pathophysiology, especially the role of gastric acid secretion can help gain a better understanding of how to effectively treat FD. Our findings, based on longer follow-up than most previous studies, confirm the major findings of E effectiveness and the relationship of E to intragastric pH, when measured at 4 and 8 weeks after the initiation of therapy.
      Johnson et al conducted a 2-way crossover study in 45 patients (H pylori–negative patients) with GERD randomized to receive either lansoprazole 30 mg QD or E 40 mg with dose increased to BID for each therapy and pH assessment at and on day 10 of therapy. They found the mean time pH > 4.0 and mean 24-hour pH were highest for E 40 mg BID followed by lansoprazole 30 mg QD, E 40 mg QD and then lansoprazole 30 mg QD.
      • Johnson D.A.
      • Stacy T.
      • Ryan M.
      • et al.
      A comparison of esomeprazole and lansoprazole for control of intragastric pH in patients with symptoms of gastro-oesophageal reflux disease.
      They concluded that the response to acid suppression varies by therapeutic regimen but E 40 mg BID provided the best control of intragastric pH compared to the other regimens in the study.
      Meineche-Schmidt et al studied 805 patients with uninvestigated dyspepsia that were considered to be acid related, and who were randomized to either 2 weeks of E 40 mg or P daily therapy. A response to therapy was observed in 68% of the E group and 44% in the P group (P < 0.00001).
      • Meineche-Schmidt V.
      • Christensen E.
      • et al.
      Randomized clinical trial: identification of responders to short-term treatment with esomeprazole for dyspepsia in primary care—a randomized, placebo-controlled study.
      The multinational Supportive Test for Acid-Related Symptoms (STARS) II study demonstrated E to be moderately effective for symptomatic improvement in patients with dyspepsia. In this study, van Zanten et al studied 1,250 patients with uninvestigated dyspepsia who underwent 1 week of E therapy and was then randomized to either E or P group for 7 weeks. Following 1 week of E therapy, the proportion of patients who responded was similar between those that received E 40 mg QD (39%) or BID (43%) (P = 0.16).
      • van Zanten S.V.
      • Wahlqvist P.
      • Talley N.J.
      • et al.
      Randomised clinical trial: the burden of illness of uninvestigated dyspepsia before and after treatment with esomeprazole—results from the STARS II study.
      After being treated for 7 weeks, there was a significant increase in the percentage of R to E than P (47% versus 34%; P < 0.001).
      Van Zanten et al performed a randomized, placebo-controlled trial in 224 adult patients with FD with moderate severity symptoms. Patients who had predominant symptoms of heartburn or regurgitation were excluded. A baseline endoscopy was performed. Patients were randomized to either E 40  mg once daily or P for 8 weeks.
      • van Zanten S.V.
      • Armstrong D.
      • Chiba N.
      • et al.
      Esomeprazole 40 mg once a day in patients with functional dyspepsia: the randomized, placebo-controlled “ENTER” trial.
      The 4-week symptom relief of E was 50.5% versus P 32.2%, P = 0.009 and 8-week symptom relief of E was 55.1% versus P 46.1%, P = 0.16. This study concluded that E was more effective than P for symptom relief at 4 weeks. They were uncertain as to why there was a difference in therapeutic gain at 8 weeks in the P group. In another study, van Zanten et al
      • van Zanten S.V.
      • Flook N.
      • Talley N.J.
      • et al.
      One-week acid suppression trial in uninvestigated dyspepsia patients with epigastric pain or burning to predict response to 8 weeks׳ treatment with esomeprazole: a randomized, placebo-controlled study.
      aimed to evaluate whether 1 week of acid suppression would help predict the response of these patients at 8 weeks of therapy. In a randomized, placebo-controlled trial of H pylori–negative patients with uninvestigated epigastric pain or burning, patients were either started on E 40 mg QD or BID for 1 week followed by E 40 mg QD or P for 7 weeks. The patients rated the severity of their symptoms on a daily basis. The response rate at the end of 1 week was 39% and 43% with E 40 mg QD and BID, respectively. At 4 weeks, the response rate was 38% for E and 25% for P. At 8 weeks, the response rate was 47% for E and 34% for P.
      • van Zanten S.V.
      • Flook N.
      • Talley N.J.
      • et al.
      One-week acid suppression trial in uninvestigated dyspepsia patients with epigastric pain or burning to predict response to 8 weeks׳ treatment with esomeprazole: a randomized, placebo-controlled study.
      They concluded that the predictive value of therapy response based on 1 week of acid suppression therapy is of limited clinical value. E provided better symptomatic control compared to P at 4 and 8 weeks of therapy.
      Talley et al
      • Talley N.J.
      • Vakil N.
      • Lauritsen K.
      • et al.
      Randomized-controlled trial of esomeprazole in functional dyspepsia patients with epigastric pain or burning: does a 1-week trial of acid suppression predict symptom response?.
      also evaluated whether 1 week of acid suppression was useful in identifying true R to E therapy. Patients were randomized to either receive E 40 mg QD or BID for 1 week followed by E 40 mg QD or P for 7 weeks. Dyspepsia symptoms such as epigastric burning sensation or pain were recorded using a 4-point Likert scale (0 = none to 3 = severe). A favorable symptom response to therapy was considered to be a symptom sum score of 1 or less on last 3 days of therapy during the first week and same score but the last 7 days at the end of 8 weeks of therapy. At the end of 1 week of therapy, the response rates were 33% for E QD, 29% for E BID and 23% for P (P = 0.002 for E group versus P). The overall response rates at week 4 were 27% with E and 26% with P (P = 0.86). The overall response rates at week 8 were 39 % for E and 33% for P. However, they concluded that response to 1-week acid suppression trial is of limited use for predicting symptom response at 8 weeks in patients with uninvestigated dyspepsia.
      • Talley N.J.
      • Vakil N.
      • Lauritsen K.
      • et al.
      Randomized-controlled trial of esomeprazole in functional dyspepsia patients with epigastric pain or burning: does a 1-week trial of acid suppression predict symptom response?.
      One of the potential downfalls of conducting clinical trials in patients with FD is the overlap that could exist with GERD.
      • Roman S.
      • Serraj I.
      • Damon H.
      • et al.
      Correlation between gastric pH and gastro-oesophageal reflux contents: ambulatory pH-impedance monitoring results.
      • Weijenborg P.W.
      • Cremonini F.
      • Smout A.J.
      • et al.
      PPI therapy is equally effective in well-defined non-erosive reflux disease and in reflux esophagitis: a meta-analysis.
      In our study, we included patients who met criteria for investigated dyspepsia further classified as functional or non–ulcer dyspepsia. Many of the studies mentioned above note that PPIs are effective at acid suppression of pH > 4 and in symptomatic relief of dyspepsia.
      None of these studies comprehensively investigate simultaneous measurement of intragastric pH and symptomatic improvement of dyspepsia with PPI therapy. Our study demonstrates that most patients who achieved a longer duration of time of pH > 4.0 also had better symptom improvement of dyspepsia with E compared to P.
      The administration of E resulted in a statistical significant increase of time pH > 4 value as compared to baseline at either the proximal or distal channel at 4 weeks (second pH test) and at 8 weeks (third pH test). The P group did not have a significant increase of time pH > 4. This is an expected finding that E does in fact cause acid suppression compared to P, specifically more at the distal gastric portion. However, the observation was that the greater the time pH > 4 correlated with symptomatic improvement of dyspepsia. This was not observed for every patient. NR (based on self-reports of symptoms) had adequate gastric acid suppression with E, but the time pH > 4 was not as long as those who responded to therapy. Therefore, the time pH > 4 did not always correlate with improvement in dyspepsia. We speculate that other factors are involved in addition to increased gastric acidity resulting in dyspepsia.
      The only exception in our data occurred in the comparison of R to P at the fourth week (second pH test) at the distal channel with pH > 4 values being increased by 47% compared to baseline value. It is known that a placebo effect can be substantial in trials of GI disorders especially dyspepsia. However, we are uncertain of the reason for such a significant increase time pH > 4 in the placebo group after 4 weeks of therapy and none after 8 weeks. One possible explanation is that this is because of a natural variation in the severity of the symptoms and is independent of treatment. Another may be that these patients might benefit by receiving increased attention, that is, by participating in the study they feel they are being treated appropriately. However, this modifier effect may apply to either group not just the placebo arm. This study was not designed to evaluate this possibility. It is unknown if there are maximum placebo effects in patients with FD, or if this response will plateau over time.
      The current treatment options for FD are still suboptimal. Unfortunately, no other treatment approach has convincingly been shown to be more efficacious. A trial of a PPI for at least 4-8 weeks is a reasonable therapy option.
      • Gwee K.A.
      • Hwang J.E.
      • Ho K.Y.
      • et al.
      In-practice predictors of response to proton pump inhibitor therapy in primary care patients with dyspepsia in an Asian population.
      PPIs are well tolerated and adverse events are minimal. Our findings support the use of PPIs for empiric therapy for patients with uninvestigated dyspepsia, as recommended by the current guidelines. An 8-week trial of acid suppression with once daily dosing of a PPI is a reasonable course of action in patients with FD who have predominant symptoms of pain or burning in the upper abdomen.
      • Weijenborg P.W.
      • Cremonini F.
      • Smout A.J.
      • et al.
      PPI therapy is equally effective in well-defined non-erosive reflux disease and in reflux esophagitis: a meta-analysis.
      The huge difference in the benefit of PPI therapy over placebo was much unexpected. This study highlights that in patients with uninvestigated dyspepsia who present with epigastric pain or burning, the burden of illness is substantial. Symptom resolution with acid-suppressive therapy is accompanied by significant improvements in quality of life and improvement in productivity compared with the placebo group.
      • Sarnelli G.
      • De Giorgi F.
      • Efficie E.
      • et al.
      Correlation between oesophageal acid exposure and dyspeptic symptoms in patients with nonerosive reflux disease.
      This study supports our hypothesis that evaluating gastric acid secretion can help gain a better understanding of how to effectively treat FD. In E patients with dyspepsia E, 40 mg QD provides profound inhibition of gastric acid secretion as reflected in the time pH > 4.0 in patients accompanied by a significant relief of symptoms. Though the positive differences between E-treated patients and placebo patients persisted throughout the study, significant improvements in symptoms and pH values were observed in the placebo patients. A significantly higher degree of gastric acid inhibition in R than NR after prolonged (8 weeks) therapy with 40 mg of E may imply that gastric acid pH monitoring could be helpful in the prediction of therapeutic response and tailoring individualized therapy. Our findings suggest that a trial P followed by E arm might be considered in a future trial.
      Additionally, it was clear that though E administered 40  mg QD is superior to placebo, in controlling symptoms and gastric acid, some FD patients may benefit from a dose of E 40 mg BID.
      These results also suggest that it would be valuable to further explore whether profound inhibition of gastric acid secretion with a more potent PPI would provide satisfactory relief of symptoms in almost all patients with FD.

      Study Strengths and Limitations

      This is the first study that we are aware of that has used a double-blind protocol design employing drug titration in patients with dyspepsia in active and placebo arms. The therapeutic phase of 16 weeks exceeds follow-up from most trials in the literature, which have typically measured outcomes at 1-8 weeks. Protocol adherence was excellent. The trial only included patients with FD without possible a number of complicating comorbidities, not requiring chronic treatment with other medications. In addition, patients were permitted to use over-the-counter (OTC) medications if their symptoms were severe. Therefore, we are unable to assess the effect that any OTC medications could have on the placebo or treatment groups, nor the effect of comorbid conditions. In an effort to develop broad treatment guidelines, further trials should address OTC use and may be designed to include patients with FD with comorbid conditions and should follow-up patients for a longer duration of therapy.

      Conclusions

      • (1)
        Although E administered 40  mg QD is superior to placebo, some patients with FD benefit from a dose of 40 mg BID.
      • (2)
        E, 40 mg QD, provides profound inhibition of gastric acid secretion as reflected in time pH > 4.0 with dyspepsia accompanied by a significant relief of symptoms.
      • (3)
        Intragastric pH monitoring before and after therapy may help address the interrelationship between symptomatic relief of dyspepsia symptoms and gastric acid secretion inhibition and could potentially help to tailor individual therapy.
      • (4)
        Though titrated E dosing appears significantly more effective regarding the symptoms and gastric action production, there appears to be a positive response in some patients with FD to monitored titrated placebo therapy.

      Acknowledgment

      The authors wish to thank Mr. Jim Bramich, MIS from AstraZeneca for his great support and encouragement.

      References

        • Brun R.
        • Kuo B.
        Functional dyspepsia.
        Therap Adv Gastroenterol. 2010; 3: 145-164
        • Lacy B.E.
        • Talley N.J.
        • Locke 3rd, G.R.
        • et al.
        Review article: current treatment options and management of functional dyspepsia.
        Aliment Pharmacol Ther. 2012; 36: 3-15
        • Robinson M.
        Dyspepsia: challenges in diagnosis and selection of treatment.
        Clin Ther. 2001; 23: 1130-1144
        • Tack J.
        • Talley N.J.
        • Camilleri M.
        • et al.
        Functional gastroduodenal disorders.
        Gastroenterology. 2006; 130: 1466-1479
        • Talley N.J.
        • Vakil N.B.
        • Moayyedi P.
        American Gastroenterological Association technical review on the evaluation of dyspepsia.
        Gastroenterology. 2005; 129: 1756-1780
        • Lacy B.E.
        Functional dyspepsia and gastroparesis: one disease or two?.
        Am J Gastroenterol. 2012; 107: 1615-1620
        • van Zanten S.V.
        • Armstrong D.
        • Chiba N.
        • et al.
        Esomeprazole 40 mg once a day in patients with functional dyspepsia: the randomized, placebo-controlled “ENTER” trial.
        Am J Gastroenterol. 2006; 101: 2096-2106
        • Thumshirn M.
        Pathophysiology of functional dyspepsia.
        Gut. 2002; 51: i63-i66
        • Moayyedi P.
        • Soo S.
        • Deeks J.
        • et al.
        Eradication of Helicobacter pylori for non–ulcer dyspepsia. Cochrane.
        Database System Review. 2005; 1: CD002096
        • Vachhani R.
        • Olds G.
        • Velanovich V.
        Esomeprazole: a proton pump inhibitor.
        Expert Rev Gastroenterol Hepatol. 2009; 3: 15-27
        • Niu X.P.
        • Yu B.P.
        • Wang Y.D.
        • et al.
        Risk factors for proton pump inhibitor refractoriness in Chinese patients with non-erosive reflux disease.
        World J Gastroenterol. 2013; 19: 3124-3129
        • McKeage K.
        • Blick S.K.
        • Croxtall J.D.
        • et al.
        Esomeprazole: a review of its use in the management of gastric acid-related diseases in adults.
        Drugs. 2008; 68: 1571-1607
        • Yang H.
        • Li J.
        • Zhao Q.
        • et al.
        Pharmacokinetics, pharmacodynamics, and safety of esomeprazole injection/infusion in healthy Chinese volunteers: a five-way crossover study.
        J Gastroenterol Hepatol. 2013; 28: 1823-1828
        • Rotman S.R.
        • Bishop T.F.
        Proton pump inhibitor use in the U.S. ambulatory setting, 2002-2009.
        PLoS One. 2013; 8: e56060
        • Reimer C.
        • Søndergaard B.
        • Hilsted L.
        • et al.
        Proton-pump inhibitor therapy induces acid-related symptoms in healthy volunteers after withdrawal of therapy.
        Gastroenterology. 2009; 137: 80-87
        • Bytzer P.
        • van Zanten S.V.
        • Mattsson H.
        • et al.
        Partial symptom-response to proton pump inhibitors in patients with non-erosive reflux disease or reflux oesophagitis—a post hoc analysis of 5796 patients.
        Aliment Pharmacol Ther. 2012; 36: 635-643
        • Johnson D.A.
        • Stacy T.
        • Ryan M.
        • et al.
        A comparison of esomeprazole and lansoprazole for control of intragastric pH in patients with symptoms of gastro-oesophageal reflux disease.
        Aliment Pharmacol Ther. 2005; 22: 129-134
        • Meineche-Schmidt V.
        • Christensen E.
        • et al.
        Randomized clinical trial: identification of responders to short-term treatment with esomeprazole for dyspepsia in primary care—a randomized, placebo-controlled study.
        Aliment Pharmacol Ther. 2011; 33: 41-49
        • van Zanten S.V.
        • Wahlqvist P.
        • Talley N.J.
        • et al.
        Randomised clinical trial: the burden of illness of uninvestigated dyspepsia before and after treatment with esomeprazole—results from the STARS II study.
        Aliment Pharmacol Ther. 2011; 34: 714-723
        • van Zanten S.V.
        • Flook N.
        • Talley N.J.
        • et al.
        One-week acid suppression trial in uninvestigated dyspepsia patients with epigastric pain or burning to predict response to 8 weeks׳ treatment with esomeprazole: a randomized, placebo-controlled study.
        Aliment Pharmacol Ther. 2007; 26: 665-672
        • Talley N.J.
        • Vakil N.
        • Lauritsen K.
        • et al.
        Randomized-controlled trial of esomeprazole in functional dyspepsia patients with epigastric pain or burning: does a 1-week trial of acid suppression predict symptom response?.
        Aliment Pharmacol Ther. 2007; 26: 673-682
        • Roman S.
        • Serraj I.
        • Damon H.
        • et al.
        Correlation between gastric pH and gastro-oesophageal reflux contents: ambulatory pH-impedance monitoring results.
        Neurogastroenterol Motil. 2007; 19: 562-568
        • Weijenborg P.W.
        • Cremonini F.
        • Smout A.J.
        • et al.
        PPI therapy is equally effective in well-defined non-erosive reflux disease and in reflux esophagitis: a meta-analysis.
        Neurogastroenterol Motil. 2012; 24 (e350): 747-757
        • Gwee K.A.
        • Hwang J.E.
        • Ho K.Y.
        • et al.
        In-practice predictors of response to proton pump inhibitor therapy in primary care patients with dyspepsia in an Asian population.
        J Clin Gastroenterol. 2008; 42: 134-138
        • Sarnelli G.
        • De Giorgi F.
        • Efficie E.
        • et al.
        Correlation between oesophageal acid exposure and dyspeptic symptoms in patients with nonerosive reflux disease.
        Eur J Gastroenterol Hepatol. 2008; 20: 264-268