Combination of proton pump inhibitor and rebamipide, a free radical scavenger, promotes artificial ulcer healing after endoscopic submucosal dissection with dissection size >40 mm.

In our previous study, the healing effect of proton pump inhibitor plus rebamipide for endoscopic submucosal dissection-related artificial ulcer smaller than 40 mm showed statistical significance. However, such effect of the combination was not yet clear for ulcers with dissected diameter more than 40 mm. The aim of this present study was to resolve this problem under sufficient statistical power, with adequate sample size. We conducted a randomized controlled study. Either the proton pump inhibitor mono-therapy or the combination therapy was prescribed for 28 days after endoscopic submucosal dissection. Eighty-seven patients were eligible for outcome evaluation. Combination therapy was significantly superior to mono-therapy, 27.8% vs 0% reached healing stage (scar stage) in cases with ulcers of dissection diameter more than 40 mm. In conclusion, the combination therapy with rebamipide was favorable regimen in patients with larger artificial ulcer after endoscopic submucosal dissection.

RCT Entities:   Population Interventions Outcomes

Introduction

Endoscopic submucosal dissection (ESD) has become popular around the world to treat patients with early gastric cancer. The current interest in ESD is regarding its application for larger tumor size. As the problematic issue in ESD procedure, delayed bleeding after ESD from artificial ulcer was focused. Gotoda et al.( reported that frequency was up to 7%. Moreover, Uedo et al.( demonstrated that the size of a tumor larger than 20 mm was an independent predictive factor for delayed bleeding. Okada et al.( demonstrated that a dissected specimen of diameter greater than 40 mm was the only significant factor associated with delayed bleeding. Expand application of ESD lead to increase the risk of delayed bleeding from artificial ulcer. Moreover, bleeding in patients discharge is fatal. Therefore, management of larger size artificial ulcer is important. Although proton pump inhibitor (PPI) or H2 receptor antagonist (H2RA) is effective for the prevention of bleeding and the healing of artificial ulcer after ESD,( it has a therapeutic limitation. In our previous study, the healing ratio in 4 weeks PPI mono-therapy was decreasing as larger as dissection specimen size: 33.3% (17/51) for specimen size <40 mm and 20.0% (2/10) for ⩾40 mm.( Administration of rebamipide, a cytoprotective agent, with PPI showed significantly superior healing effect on ulcers with larger dissected size of 20 to 40 mm as compared with PPI mono-therapy. Such effect of the combination was suggested also for ulcers with a diameter greater than 40 mm: 75% (6/8) by the combination therapy and 20% (2/10) by the mono-therapy. However, there was no statistical significance (p = 0.058 by Fischer’s exact test) because of less statistical power due to the small number of patients with such dissection size. Therefore, we conducted the following study to prove the healing effect of the combination therapy for artificial ulcer with a diameter greater than 40 mm with adequate sample size calculated from our previous data.

Methods

Subjects

This study was conducted from 2007 to 2010 at Gifu University Hospital Endoscopy Center and Gifu Municipal Hospital. The study protocol was approved by the Institutional Review Board for Human Research at each hospital. Inclusion criteria were all patients with gastric adenoma or early cancer who attended our hospitals for the purpose of ESD with a predicted resection size over 20 mm. Exclusion criteria were those with repeat bloody vomitus, major organ failure such as cardiac and pulmonary complications, or drug allergy. Those receiving concomitant administration of anticoagulants and anti-inflammatory drugs were also excluded. Written informed consent was obtained from all participants.

Study drugs and treatment schedule

Patients were admitted one day before ESD and hospitalized for at least 7 days post-ESD. All patients received an intravenous administration of omeprazole on the first two days after ESD. Participants then took either PPI (omeprazole 20 mg/day, lansoprazole 30 mg/day or rabeprazole 10 mg/day) alone or PPI plus 100 mg rebamipide 3 times/day for the following 26 days.

Endpoints

The primary endpoint was to evaluate the artificial ulcer healing ratio of the combination therapy compared with the PPI mono-therapy in all patients after ESD. The secondary endpoint was to evaluate the effect of the combination therapy in patients with a dissection size greater than 40 mm.

Outcome evaluation

Artificial ulcer healing was evaluated by endoscopy on the 28th day. The Sakita and Miwa classification, the scar stage was defined as healing of ulcer, was an evaluation scale for the ulcer stage.( The dissection size was measured by pinning the specimen flat on a rubber plate. Endoscopic evaluation for the ulcer stage on the 28th day was judged by a representative blinded gastroenterologist, H.A. in Gifu University Hospital.

Calculation of the sample size

In our previous study on ESD ulcers ⩾20 mm,( 68% of patients in the combination group progressed to scar stage while only 35% progressed in the PPI mono-therapy group (33% improvement, p = 0.011). Therefore, at least a total of 35 cases were calculated as required sample size in one arm with α-error <0.05 and β-error <0.2. Regarding cases with ulcer greater than 40 mm, 6 of 8 (75%) patients in the combination group progressed to scar stage as compared with 2 of 10 (20%) in the PPI mono-therapy group. These data indicate that at least 12 cases with such ulcer size were needed in one arm. In addition, only 30% of all patients had this artificial ulcer size in the previous study. Thus, the required number of total cases in one arm was 40. By adding 10% possible ineligibility in ulcer size and 10% dropouts, 48 was set as the target sample size in one arm.

Statistical analysis

Baseline demographic and clinicopathologic data were compared using the chi-square test or Mann-Whitney U test. We used the chi-square test or Fisher’s exact test with a 95% confidence interval to compare the endpoint ulcer stage between the two groups. The significance of difference was declared with a p value of less than 0.05. All statistical analyses were performed using JMP ver. 8.0.2 software (SAS Institute, Cary, NC).

Results

Patient flow

As shown in Fig. 1, a total of 100 patients were eligible and randomly assigned to the two groups. ESD was successfully performed and was judged as curative histopathologically in all cases. In 3 patients in the combination therapy group and in 5 cases in the PPI mono-therapy group, the dissected specimen size did not reach 20 mm. Lost to follow-up were 3 in the combination group, and 2 in the mono-therapy group.

Fig. 1

Patient flow.

Baseline demographic and clinicopathologic data

The baseline demographic and clinicopathologic data were not different between the 2 groups, as shown in Table 1.

Table 1

Baseline demographic and clinicopathologic data of PPI alone and PPI + rebamipide combination groups

The effect of combination therapy

The ratio of S1 stage at 28 days after ESD was 6 of 42 (14.3%) in the mono-therapy group, and 20 of 45 (44.4%) in the combination group, respectively (absolute difference: 30.1, 95% CI: 12.1–48.1) (p = 0.0021) (Fig. 2). In cases with dissection diameter greater than 40 mm, the rate of S1 stage was 0 of 16 (0%) in the mono-therapy group, and 5 of 18 (27.8%) in the combination group (absolute difference: 27.8, 95% CI: 7.0–48.4) (p = 0.0465) (Fig. 2). For ulcers between 20 and 40 mm, the rates were 23.0% (6/26) and 55.6% (15/27), respectively (absolute difference: 32.6, 95% CI: 7.7–57.2) (p = 0.0244) (Fig. 2).

Fig. 2

The ratio of Scar stage at 28 days after ESD. CI: confidential interval, chi-square test.

Discussion

In our previous data, the combination therapy was useful as compared with PPI mono-therapy for ESD-ulcer healing. However, the effect of the combination was not significant on ulcers larger than 40 mm,( because the statistical power was insufficient due to the small sample size of this ulcer size. In order to address this concern, we analyzed the proportion of patients with larger ulcers in the total ESD cases referring to our previous data, and designed this study with 48 cases as the target for one arm. As a result, healing ratio of artificial ulcer rose in the combination therapy group by an absolute difference of 30.1% (95% CI: 12.1–48.1) in all cases. Moreover, the ratio was also improved by a similar absolute difference of 27.8% (95% CI: 7.0–48.4) in ulcers larger than 40 mm.

Although PPI is the standard of care for ESD-related artificial ulcer and bleeding after ESD, Fujiwara et al.( reported that PPIs and acid-suppressive agents alone have a limited effect in low acid secretion, such as severe atrophic condition in the gastric mucosa. Terano et al.( reported that 7 weeks of rebamipide treatment without acid-suppressive agents promoted gastric ulcer healing after eradication therapy compared with a placebo treatment. Further evidence of the combination therapy for ESD ulcer has recently been added by Fujiwara et al.(

Rebamipide promote healing of mucosal injury by inducing the up-regulation of EGF and EGF receptors,( and by phosphorylation of extracellular signal-regulated kinase (ERK) and activation of Rho kinase.( Moreover, rebamipide maintained gastric mucosal integrity by increasing prostaglandin concentration.( Patients taking rebamipide concomitantly with PPI or H2RA had mucosal injury less frequently than those taking acid suppressants plus other mucoprotective drugs.( Our study design employed 4 weeks of treatment after ESD according to the report by Uedo et al.( that most bleeding after ESD occurred within 28 days. Therapeutic strategy for artificial ulcer healing within 4 weeks may thus have a clinical significance to prevent bleeding after ESD. Major bleeding in both vomiting and stool was not actually observed in this study.

In conclusion, combination therapy for artificial ulcer healing within 4 weeks might be useful to prevent ESD-related complications in patients with larger dissection area.