Journal of Gastroenterology

, Volume 45, Issue 12, pp 1193–1200 | Cite as

Risk factors for relapse of erosive GERD during long-term maintenance treatment with proton pump inhibitor: a prospective multicenter study in Japan

  • Kazuma Fujimoto
  • Michio Hongo
  • The Maintenance Study Group
Original Article—Alimentary Tract

Abstract

Background

Despite low gastric acid secretion levels among elderly people and Helicobacter pylori-positive individuals in Japan, many patients suffer from endoscopic relapse of erosive gastroesophageal reflux disease (GERD) during standard-dose proton pump inhibitor (PPI) maintenance therapy. We aimed to investigate the relapse rate and risk factors for relapse during long-term PPI therapy in a prospective multicenter study.

Methods

Patients diagnosed endoscopically as having erosive GERD according to the Los Angeles (LA) classification, with remission under PPI medication, and without mucosal damage in the esophagus, were enrolled and took rabeprazole 10 mg/day, the standard dose in Japan, for up to 104 weeks, with endoscopy at weeks 24, 52, 76, and 104; erosive GERD with LA classification grade A, B, C, or D was defined as relapse.

Results

The baseline status of the 191 cases analyzed was: female (34.6%), ≥65 years old (50.8%), H. pylori-positive (40.8%), body mass index (BMI) ≥25 kg/m2 (35.6%), and hiatus hernia (79.6%). Relapse occurred by week 104 in 21 cases (11%; 12 females, 9 males). Risk factors were hiatus hernia; severe past erosive GERD (grade C or D); H. pylori-negative; no mucosal atrophy; nonsmoking; and being female and <150 cm in height.

Conclusions

This study revealed the significant risk factors that might be monitored during long-term maintenance therapy to prevent relapse of GERD.

Keywords

Reflux esophagitis Hiatus hernia Helicobacter pylori Los Angeles classification 

Introduction

Recently, due to the westernization of dietary habits, the aging of society, and increased gastric acid secretion [1], there has been an increase in the number of patients with gastroesophageal reflux disease (GERD) in Japan [2]. In patients with erosive GERD, proton pump inhibitors (PPIs), which strongly inhibit gastric acid secretion, are used to heal the gastric mucosal injury caused by the esophageal reflux of gastric acid. PPIs are used at standard or half the standard dosage for maintenance therapy even after GERD has been cured.

The basal acid output in patients with erosive GERD is significantly higher than that in healthy individuals [3], and even during remission, gastric acid secretion levels in patients with recurrent GERD persist at levels similar to those present at the time of GERD onset [4]. Moreover, based on a report indicating a correlation between the GERD cure rate and the pH >4 holding time [5], it is believed that strong inhibition of gastric acid secretion during maintenance therapy improves patient quality of life (QOL).

On the other hand, relapse has occasionally been observed endoscopically during proton pump inhibitor (PPI) maintenance therapy in some patients with GERD who were initially cured by PPI therapy [6, 7]. In addition to an insufficient inhibitory effect on gastric acid secretion, various other factors, including complications from esophageal hiatus hernia and decreased lower esophageal sphincter (LES) pressure may be responsible for the relapse.

Due to the higher prevalences of elderly people and Helicobacter pylori-positive individuals in Japan compared to those in other countries, the relapse rates and risk factors may differ from those in other countries. Accordingly, we conducted a prospective multicenter clinical trial and investigated the relapse rates and risk factors for the relapse of erosive GERD in patients currently undergoing long-term PPI maintenance therapy.

Subjects, materials, and methods

Study subjects

This study included patients ≥20 years old with erosive GERD who were taking or who required PPI maintenance therapy. Patients having been cured of erosive GERD endoscopically at the time of enrollment were enrolled. Patients were enrolled regardless of whether or not they had subjective symptoms. Patients were excluded if they had perforation of the esophagus, pyloric stenosis, or esophageal varices; if they were regarded as candidates for surgical therapy; if they had a history of upper gastrointestinal tract surgery or vagotomy; or if they had complications of severe cardiac, hematologic, renal, hepatic, respiratory, or malignant disorders.

Study design

This study was a prospective, multicenter trial conducted in 27 institutions in Japan. Patients enrolled in the trial took a PPI, rabeprazole, 10 mg/day (standard dosage in Japan), once daily for 104 weeks. The following medications were prohibited during the study period: other PPIs, H2 receptor antagonists, selective muscarinic receptors, and prokinetics. H. pylori eradication was also prohibited during the study period.

The study was conducted in accordance with the Declaration of Helsinki. All patients gave their written informed consent before entering the study. Before the conclusion of 52 weeks of treatment with the investigational drug, the attending physician discussed continuing participation in the trial with the patient, and patients who provided their written informed consent were given the investigational drug for an additional 52 weeks. The study protocol and the informed consent form were approved by the institutional review boards at each institution.

Study visits and evaluations

Subject characteristics before administration were examined. The presence of hiatus hernia was endoscopically confirmed according to the Makuuchi classification [8, 9]. Patients visited the hospital every 4 weeks after the start of the study treatment, and their compliance with the study treatment was checked at each visit by counting the remaining drugs that patients brought. The consumed drug rate was categorized as ≥90, ≥75 to >90, ≥50 to >75, and >50%. Endoscopy was performed 5 times, during visits at weeks 0, 24, 52, 76, and 104. A relapse was defined as endoscopic findings of grade A, B, C, or D based on the Los Angeles classification.

Evaluation of gastric atrophy

During the endoscopy of each visit, endoscopic specialists rated the degree of gastric mucosal atrophy according to the Kimura-Takemoto classification [10]. Kimura and Takemoto have reported that the endoscopic border of atrophy lies between the antrum and fundic gland areas, and it can be recognized endoscopically by the color and elevation of the mucosa. Atrophic patterns are classified as either a closed type (C-0, C-1, C-2, and C-3) or an open type (O-1, O-2, O-3, and O-p) based on where this border exists.
  • C-0: no atrophy

  • C-1: atrophy in pylorus

  • C-2: atrophy extending to the incisura angularis

  • C-3: atrophic border from pyloric greater curvature to lesser curvature crossing anterior wall

  • O-1: atrophic border between lesser curvature and anterior wall

  • O-2: atrophic border within anterior wall

  • O-3: atrophic border between anterior wall and greater curvature

  • O-p: atrophy in entire stomach

Serum gastrin analysis

Blood samples were obtained when patients were in a fasting state, at weeks 0, 24, 52, 76, and 104, and sent to an outside contract laboratory (SRL Medisearch, Tokyo, Japan). Serum gastrin levels (SGLs) were measured using a radioimmunoassay kit (Gastrin-RIA kit II; SRL, Tokyo, Japan), with normal values considered as ≤200 pg/mL.

H. pylori tests

Blood samples were collected from patients during the visit at week 0, and the serum was analyzed for H. pylori IgG antibody by SRL Medisearch. Serum immunoglobulin G antibodies to H. pylori were measured using commercial enzyme immunosorbent assay (EIA) tests (JHM-CAP; Scimedx, Denville, NJ, USA). EIA values (EVs) of ≥2.3 and <2.3 were judged to be positive and negative, respectively.

Statistical analysis

For each of the demographic and other baseline characteristics, the results were tabulated by gender, as well as for the full analysis set. The mean values for height and body weight were graphed by age and by gender on bar graphs. Next, to conduct an exploratory investigation of the risk factors for the relapse of erosive GERD, the hazard ratios (HRs) for each risk factor were calculated using a univariate Cox proportional hazard model. Hazard ratios were also calculated for age, height, body weight, and body mass index (BMI) by applying the univariate Cox proportional hazard model tabulated by gender. To explore how the relapse of erosive GERD was affected by the SGL, SGLs were graphed over time using mean values ± standard deviation (SD) from visits at weeks 0, 24, 52, 76, and 104 in both relapsed and non-relapsed patients.

Results

Demographic and baseline characteristics

Table 1 shows the demographic and baseline characteristics, by gender, of the 191 patients who underwent endoscopic examination after being given the investigational drug. Three-quarters of the women were elderly (≥65 years of age), and at least half had a height of <150 cm. Regarding clinical symptoms, 151 patients had no symptoms at all at the enrollment and 40 patients had some symptoms. Of these 40 patients, two had moderate symptoms while the rest had mild symptoms. Figure 1 shows the age, mean body weight, and mean height distributions by gender. In men, mean height and mean body weight showed a tendency to decrease with age, while in women, although mean height decreased, mean body weight remained essentially the same.
Table 1

Patients’ demographic and baseline characteristics

 

Male (n = 125)

Female (n = 66)

Total (n = 191)

Age, years (mean ± SD)

59.4 ± 12.9

69.8 ± 10.8

63.0 ± 13.2

 <65

78 (62.4)

16 (24.2)

94 (49.2)

 ≥65

47 (37.6)

50 (75.8)

97 (50.8)

Height (mean ± SD)

166.3 ± 6.5

148.7 ± 6.3

160.2 ± 10.6

 <150 cm

0 (0.0)

37 (56.1)

37 (19.4)

 ≥150 to <165 cm

55 (44.0)

29 (43.9)

84 (44.0)

 ≥165 cm

70 (56.0)

0 (0.0)

70 (36.6)

Weight (mean ± SD)

67.6 ± 9.3

51.9 ± 8.0

62.1 ± 11.6

 <50 kg

2 (1.6)

29 (43.9)

31 (16.2)

 ≥50 to <65 kg

52 (41.6)

33 (50)

85 (44.5)

 ≥65 kg

71 (56.8)

4 (6.1)

75 (39.3)

BMI (mean ± SD)

24.4 ± 2.6

23.4 ± 3.1

24.0 ± 2.8

 <25 kg/m2

79 (63.2)

44 (66.7)

123 (64.4)

 ≥25 kg/m2

46 (36.8)

22 (33.3)

68 (35.6)

H. pylori infection

 Negative

78 (62.4)

35 (53.0)

113 (59.2)

 Positive

47 (37.6)

31 (47.0)

78 (40.8)

SGL (median, min–max) in H. pylori-negative patients

110 (17–640)

170 (20–590)

120 (17–640)

 <200 pg/mL

62 (79.5)

19 (54.3)

81 (71.7)

 ≥200 to <400 pg/mL

14 (18.0)

10 (28.6)

24 (21.2)

 ≥400 pg/mL

2 (2.6)

6 (17.1)

8 (7.1)

SGL (median, min–max) in H. pylori-positive patients

170 (21–890)

220 (74–810)

205 (21–890)

 <200 pg/mL

26 (55.3)

11 (35.5)

37 (47.4)

 ≥200 to <400 pg/mL

15 (31.9)

12 (38.7)

27 (34.6)

 ≥400 pg/mL

6 (12.8)

8 (25.8)

14 (18.0)

Mucosal atrophy

 Negative (C-0)

33 (26.4)

12 (18.2)

45 (23.6)

 Positive (C-1 to O-p)

92 (73.6)

54 (81.8)

146 (76.4)

Severity of past erosive GERD

 Grade A

49 (39.2)

28 (42.4)

77 (40.3)

 Grade B

48 (38.4)

23 (34.8)

71 (37.2)

 Grade C

21 (16.8)

12 (18.2)

33 (17.3)

 Grade D

6 (4.8)

2 (3.0)

8 (4.2)

 Not classified

1 (0.8)

1 (1.5)

2 (1.0)

Type of PPI used before trial

 Rabeprazole

65 (52.0)

29 (43.9)

94 (49.2)

 Lansoprazole

41 (32.8)

24 (36.4)

65 (34.0)

 Omeprazole

19 (15.2)

13 (19.7)

32 (16.8)

Smoking

 No

92 (73.6)

59 (89.4)

151 (79.1)

 Yes

33 (26.4)

7 (10.6)

40 (20.9)

Hiatus hernia

 Negative

25 (20.0)

14 (21.2)

39 (20.4)

 Positive

100 (80.0)

52 (78.8)

152 (79.6)

SD standard deviation, BMI body mass index, SGL serum gastrin level, H. pyloriHelicobacter pylori, PPI proton pump inhibitor, GERD gastroesophageal reflux disease

Fig. 1

Patient background (height and weight based on age)

Treatment compliance

Most subjects maintained good compliance with the treatment. During the administration period, 146 patients maintained a compliance rate of ≥90% at all evaluation dates every 4 weeks, 39 patients fell into the ≥75 to >90% compliance rate at least once at evaluation dates every 4 weeks, 5 patients showed compliance rates of ≥50 to >75%, and one patient showed a compliance rate of >50%. Among the 21 relapsed patients, 16 patients (76.2%) maintained a compliance rate of ≥90% on all evaluation dates. Among the 170 non-relapsed patients, 130 patients (76.5%) maintained a compliance rate of ≥90%. No significant difference existed between the relapsed group and non-relapsed group regarding compliance.

Patients experiencing relapse and risk factors for relapse

Table 2 lists the 21 patients (11.0%) in whom endoscopically verified relapse occurred during 104 weeks of treatment with the investigational drug. All patients had esophageal hernias as complications. Table 3 shows endoscopic relapse rates at week 104 calculated using Kaplan–Meier’s estimator and HRs calculated using the Cox proportional hazard model. The relapse rate was significantly higher in women than in men (HR = 2.606; 95% confidence interval [CI], 1.098–6.185); relapse rates were also significantly higher when past erosive GERD severity was “grade C or D” versus “grade A or B” (HR = 2.937; 95% CI, 1.236–6.982). The relapse rate was significantly higher in H. pylori-negative patients (HR = 0.313; 95% CI, 0.105–0.930); the relapse rate was also significantly higher in gastric mucosal atrophy-negative patients (HR = 0.366; 95% CI, 0.154–0.869). Table 4 shows the results of an investigation into the relationship between relapse and the background factors of age, height, and body weight, tabulated by gender. Among females, the relapse rate was significantly higher in those with height <150 cm than in those with height ≥150 cm (HR = 0.215; 95% CI, 0.047–0.981).
Table 2

Patients with relapse

No.

Gender

Age (years)

Height (cm)

Weight (kg)

Number of relapses in past 2 years

Severity of past erosive GERD

Time until relapse (weeks)

Severity at relapse

1

Female

74

148

53

6

Grade A

24

Grade A

2

Female

62

154

60

2

Grade C

24

Grade A

3

Male

66

157

64.5

2

Grade A

24

Grade A

4

Male

65

168

73

3

Grade C

24

Grade A

5

Male

51

170

70

3

Grade A

52

Grade A

6

Female

83

149.6

60.5

1

Grade D

52

Grade D

7

Female

82

147

48

0

Grade C

24

Grade C

8

Female

76

147.5

54.6

3

Grade A

8

Grade A

9

Female

68

149.9

69.1

2

Grade B

104

Grade B

10

Female

74

144.1

48.1

1

Grade A

52

Grade A

11

Female

74

147

49.4

1

Grade A

24

Grade A

12

Male

45

173

81

3

Grade B

24

Grade A

13

Male

60

163

60

3

Grade C

76

Grade A

14

Male

72

153

59

1

Grade C

76

Grade A

15

Female

64

140

48

3

Grade B

24

Grade B

16

Male

75

160.8

76

3

Grade B

24

Grade B

17

Female

69

151.1

57.1

2

Grade C

52

Grade B

18

Female

56

144.7

63.5

2

Grade C

104

Grade A

19

Male

25

176.6

59.5

2

Grade B

104

Grade B

20

Male

70

158.1

65

3

Grade C

24

Grade A

21

Female

69

141.5

46

2

Grade A

52

Grade A

Table 3

Hazard ratios for relapse of erosive GERD

 

Relapse rate (%)

Hazard ratio (95% CI)

Gender

 Male

8.5

2.606 (1.098–6.185)*

 Female

20.5

Age (years):

 <65

9.1

2.087 (0.841–5.176)

 ≥65

16.3

Baseline SGL (pg/mL)

 <200

13.9

 

 ≥200 to <400

13.6

1.026 (0.394–2.670)

 ≥400

5.6

0.340 (0.045–2.585)

Helicobacter pylori infection

 Negative

17.8

0.313 (0.105–0.930)*

 Positive

5.8

Smoking

 No

16.1

 Yes

0.0

 

Baseline mucosal atrophy

 Negative (C-0)

23.7

0.366 (0.154–0.869)*

 Positive (C-1 to O-p)

9.5

Hiatus hernia

 Negative

0.0

 Positive

15.7

 

Severity of past erosive GERD

 Grade A or B

9.1

2.937 (1.236–6.982)*

 Grade C or D

27.0

* Statistical significance, CI confidence interval

Table 4

Hazard ratios for relapse of erosive GERD by gender

 

Male

Female

Non-relapse rate (%)

Hazard ratio (95% CI)

Non-relapse rate (%)

Hazard ratio (95% CI)

Age (years)

 <65

93.6

 

78.8

 

 ≥65

88.2

2.229 (0.597–8.321)

79.7

1.062 (0.287–3.929)

Height (cm)

 <150

 

68.2

 

 ≥150

  

92.8

0.215 (0.047–0.981)*

 <165

89.0

 

 

 ≥165

93.2

0.570 (0.153–2.126)

  

Weight (kg)

 <50

 

81.7

 

 ≥50

  

77.9

1.065 (0.338–3.356)

 <65

89.9

 

 

 ≥65

92.7

0.902 (0.242–3.361)

  

BMI (kg/m2)

 <25

94.8

 

85.7

 

 ≥25

86.1

3.548 (0.887–14.194)

68.0

2.056 (0.663–6.377)

* Statistical significance

Serum gastrin levels in relapsed patients

Changes in the SGLs in the relapsed and non-relapsed patients, based on H. pylori infection status, are presented in Fig. 2. The SGL in the relapsed patients tended to be lower than that in the non-relapsed patients, but the difference was not significant. Number of patients evaluated at each evaluation period are shown in Table 5.
Fig. 2

Changes in the serum gastrin levels (medians). H. pylori, Helicobacter pylori

Table 5

The numbers of patients at each evaluation period

 

Week 0 

Week 24 

Week 52

Week 76 

Week 104 

H. pylori-negative non-relapse

96

92

82

74

72

H. pylori-positive non-relapse

17

16

7

4

2

H. pylori-negative relapse

74

70

70

59

57

H. pylori-positive relapse

4

4

2

1

1

Adverse events

During the 2-year testing period, 183 patients (95.8%) reported some adverse events, regardless of whether or not the adverse event was due to the tested medicine. The main adverse events were classified in large categories such as infections, gastrointestinal disorders, and so on, as indicated in Table 6. Most of the adverse events might not have been due to rabeprazole.
Table 6

Main adverse events during the two-year follow-up period

Category

n

Infections and infestations

133

Gastrointestinal disorders

111

Musculoskeletal, connective tissue, and bone disorders

68

Injury, poisoning, and procedural complications

46

Skin and subcutaneous tissue disorders

44

Respiratory, thoracic, and mediastinal disorders

42

Nervous system disorders

38

Investigations

36

Discussion

There have been many reports of studies of long-term PPI administration for GERD in Europe and the United States [11, 12]. However, there have been virtually no long-term studies in Japan, and this is the first study to investigate GERD relapse rates and prognostic factors over a 2-year period. In European and United States studies [7], 4 factors were associated with a high risk of relapse: pre-treatment severity, young age, non-smoking, and moderate/severe regurgitation. In another study [6], it was suggested that the “severity of subjective symptoms prior to GERD treatment” was a prognostic factor, but no correlation between relapse and “H. pylori infection status” was observed. The results of univariate analysis in the present study indicated that the following conditions were associated with a high risk of relapse in Japanese subjects: hiatus hernia, past severity of erosive GERD (grade C or D), H. pylori-negative, no gastric mucosal atrophy, non-smoking, and women with height <150 cm. It was not possible to include the above risk factors in the multivariate model, because none of the patients with smoking habits or without hiatus hernia in this study had relapsed erosive GERD.

In patients with esophageal hiatus hernia, it has been reported that gastric acid that has refluxed into the esophagus stalls in the hernia sac and repeatedly flows back into the esophagus, and that the clearance of gastric acid from the esophagus is low [13], which is why it is believed that relapse occurs more readily in patients with esophageal hiatus hernia than in patients without esophageal hiatus hernia. In patients with grade C or D GERD, acid reflux into the esophagus has continued for a longer time than in patients with grade A or B symptoms, in which the mucosal damage to the esophagus is mild. The pH is lower, and the conditions are more acidic in H. pylori-negative patients than in H. pylori-positive patients [14, 15]. Moreover, in patients with no atrophy, there are more gastric mucosal cells, including parietal cells that produce gastric acid, than in patients with atrophy. As a result, in patients with either “previous GERD severity: grade C or D” or “no gastric mucosal atrophy,” or in “H. pylori-negative” patients, it is believed that acid reflux into the esophagus remains at a consistently higher level than in patients without these conditions, which makes relapse more likely to occur. A reason that non-smoking was identified as a risk factor for relapse was also discussed previously [7]. It is assumed that while patients with severe GERD will stop smoking, the need to stop smoking is not as urgent for patients with mild symptoms, and, if this is the case, GERD symptoms will be milder in smokers, and, thus, relapse will be less likely to occur.

It is likely that women with a height of <150 cm have a hump back or round back, a typical body form among Japanese people; however, this was not examined in the present study, so it is impossible to be certain in this case. A hunchback decreases height due to curvature of the back, and it is frequently found in elderly Japanese women engaged in agricultural work. Moreover, this posture increases abdominal pressure and compresses the stomach, a condition in which reflux of gastric acid into the esophagus is more likely to occur [16].

The exact reason why serum gastrin levels (SGLs) at the time of relapse were not elevated in relapsed patients is not known. In general, SGLs increase during PPI treatment due to a negative feedback mechanism caused by the suppression of acid secretion. In short, the fact that SGLs were not elevated suggests that gastric acid secretion was not being sufficiently suppressed, but because even relapsed patients were taking the study treatment at a standard dosage (rabeprazole, 10 mg), there may be another reason for this finding.

Given the assumed risk of relapse occurring when these relapse factors are observed during maintenance therapy with a standard dosage of PPI after GERD has been cured, it may be necessary to take steps (increasing PPI dosage) to reinforce the suppression of acid secretion.

Notes

Acknowledgments

We would like to thank five members of the staff of Eisai Co., Ltd. (Yasuhiro Marukawa, Toshihisa Arai, Kazutaka Matsuo, Toshiyuki Tamai, and Yukinori Sakata) for their kind assistance in preparing the manuscript.

Conflict of interest statement

This study was funded by Eisai Co., Ltd. Kazuma Fujimoto and Michio Hongo are members of advisory committees of Eisai Co., Ltd.

References

  1. 1.
    Kinoshita Y, Kawanami C, Kishi K, Nakata H, Seino Y, Chiba T. Helicobacter pylori independent chronological change in gastric acid secretion in the Japanese. Gut. 1997;41:452–8.CrossRefPubMedGoogle Scholar
  2. 2.
    Miyamoto M, Haruma K, Kuwabara M, Nagano M, Okamoto T, Tanaka M. High incidence of newly developed gastroesophageal reflux disease in the Japanese community: a 6-year follow-up study. J Gastroenterol Hepatol. 2008;23:393–7.CrossRefPubMedGoogle Scholar
  3. 3.
    Collen MJ, Johnson DA, Sheridan MJ. Basal acid output and gastric acid hypersecretion in gastroesophageal reflux disease. Correlation with ranitidine therapy. Dig Dis Sci. 1994;39:410–7.CrossRefPubMedGoogle Scholar
  4. 4.
    Collen MJ, Strong RM. Comparison of omeprazole and ranitidine in treatment of refractory gastroesophageal reflux disease in patients with gastric acid hypersecretion. Dig Dis Sci. 1992;37:897–903.CrossRefPubMedGoogle Scholar
  5. 5.
    Bell NJ, Burget D, Howden CW, Wilkinson J, Hunt RH. Appropriate acid suppression for the management of gastro-oesophageal reflux disease. Digestion. 1992;51(Suppl 1):59–67.CrossRefPubMedGoogle Scholar
  6. 6.
    Hatlebakk JG, Berstad A. Prognostic factors for relapse of reflux oesophagitis and symptoms during 12 months of therapy with lansoprazole. Aliment Pharmacol Ther. 1997;11:1093–9.CrossRefPubMedGoogle Scholar
  7. 7.
    Carlsson R, Galmiche JP, Dent J, Lundell L, Frison L. Prognostic factors influencing relapse of oesophagitis during maintenance therapy with antisecretory drugs: a meta-analysis of long-term omeprazole trials. Aliment Pharmacol Ther. 1997;11:473–82.CrossRefPubMedGoogle Scholar
  8. 8.
    Makuuchi H. Clinical study of esophageal hiatal hernia: diagnostic criteria and degree classification of hiatal hernia. Jpn J Gastroenterol. 1982;79:1557–66 (in Japanese with English abstract).Google Scholar
  9. 9.
    Kusano M, Hashizume K, Ehara Y, Shimoyama Y, Kawamura O, Mori M. Size of hiatus hernia correlates with severity of kyphosis, not with obesity, in elderly Japanese women. J Clin Gastroenterol. 2008;42(4):345–50.CrossRefPubMedGoogle Scholar
  10. 10.
    Kimura K, Takemoto T. An endoscopic recognition of the atrophic border and its significance in chronic gastritis. Endoscopy. 1969;3:87–97.CrossRefGoogle Scholar
  11. 11.
    Thjodleifsson B, Rindi G, Fiocca R, Humphries TJ, Morocutti A, Miller N, et al. A randomized, double-blind trial of the efficacy and safety of 10 or 20 mg rabeprazole compared with 20 mg omeprazole in the maintenance of gastro-oesophageal reflux disease over 5 years. Aliment Pharmacol Ther. 2003;17:343–51.CrossRefPubMedGoogle Scholar
  12. 12.
    Caos A, Breiter J, Perdomo C, Barth J. Long-term prevention of erosive or ulcerative gastro-oesophageal reflux disease relapse with rabeprazole 10 or 20 mg vs. placebo: results of a 5-year study in the United States. Aliment Pharmacol Ther. 2005;22:193–202.CrossRefPubMedGoogle Scholar
  13. 13.
    Emerenziani S, Habib FI, Ribolsi M, Caviglia R, Guarino MP, Petitti T, et al. Effect of hiatal hernia on proximal oesophageal acid clearance in gastro-oesophageal reflux disease patients. Aliment Pharmacol Ther. 2006;23:751–7.CrossRefPubMedGoogle Scholar
  14. 14.
    Shimatani T, Moriwaki M, Xu J, Tazuma S, Inoue M. Acid-suppressive effects of rabeprazole: comparing 10 mg and 20 mg twice daily in Japanese Helicobacter pylori-negative and -positive CYP2C19 extensive metabolisers. Dig Liver Dis. 2006;38:802–8.CrossRefPubMedGoogle Scholar
  15. 15.
    Yamaguchi K, Iwakiri R, Hara M, Kikkawa A, Fujise T, Ootani H, et al. Reflux esophagitis and Helicobacter pylori infection in patients with scleroderma. Intern Med. 2008;47:1555–9.CrossRefPubMedGoogle Scholar
  16. 16.
    Iwakiri R, Fujimoto K. Importance of vertebral fracture and body mass index in the pathogenesis of gastroesophageal reflux disease. Intern Med. 2008;47:1551–3.CrossRefPubMedGoogle Scholar

Copyright information

© Springer 2010

Authors and Affiliations

  • Kazuma Fujimoto
    • 1
  • Michio Hongo
    • 2
  • The Maintenance Study Group
  1. 1.Department of Internal MedicineSaga Medical SchoolSagaJapan
  2. 2.Departments of Comprehensive Medicine and Psychosomatic MedicineTohoku University HospitalSendaiJapan

Personalised recommendations