AIM: To compare the therapeutic effects of proton pump inhibitors (PPI) and histamine 2 receptor antagonists (H2RA) on gastroduodenal ulcers under continuous use of low-dose aspirin.

METHODS: Sixty patients who had a gastroduodenal ulcer on screening endoscopy but required continuous use of low-dose aspirin were randomly assigned to receive PPI (lansoprazole 30 mg, n = 30) or H2RA (famotidine 40 mg or if famotidine had been administered before assignment, ranitidine 300 mg, n = 30). The therapeutic effects were evaluated by endoscopy after 8-wk treatment. The presence or absence of Helicobacter pylori (H. pylori) was determined by urea breath test before treatment. Abdominal symptoms were compared with the gastrointestinal symptom rating scale (GSRS) questionnaire before and after treatment.

RESULTS: Twenty-six patients in the PPI group and 26 patients in the H2RA group, excluding dropouts, were analyzed. There were no significant differences in median age, sex, underlying disease, smoking status, H. pylori infection, prevalence of ulcers before treatment, and lesion site between the two groups. The therapeutic effects were endoscopically evaluated as healed in 23 patients (88.5%) and not healed in 3 patients in the PPI group and as healed in 22 patients (84.6%) and not healed in 4 patients in the H2RA group. Abdominal symptoms before treatment were uncommon in both groups; the GSRS scores were not significantly reduced after treatment as compared with before treatment.

CONCLUSION: The healing rate of gastroduodenal ulcers during continuous use of low-dose aspirin was greater than 80% in both the PPI group and the H2RA group, with no significant difference between the two groups.

As the population of Japan ages, the use of low-dose aspirin has increased to prevent cerebral and myocardial infarction. Many reports from Japan and overseas showing evidence that low-dose aspirin is useful for preventing thrombosis have been published. However, low-dose aspirin use raises concern about its adverse effects such as gastrointestinal mucosal injury[1,2]. Low-dose oral aspirin (300 mg or less) is reportedly associated with a 2.6- to 3.0-fold increased risk of ulcers[3] and a 1.59-fold increased risk of gastrointestinal bleeding[4]. The prevalence of low-dose (200 mg or less) aspirin-associated gastroduodenal ulcers was 11.9% to 15.7% in Japanese patients treated for ischemic heart disease[5], and another case-control study has shown that low-dose aspirin is associated with a 5.5-fold increased risk of gastrointestinal bleeding[6]. However, low-dose aspirin is administered for the purpose of secondary prevention of cardiovascular events, and because drug suspension due to gastrointestinal injury would increase the risk of thrombosis, it is frequently difficult to discontinue the use of low-dose aspirin. In fact, when long-term users of low-dose aspirin suspended the use of the drug, the risk of thrombosis increased[7,8]. Therefore, gastrointestinal injury should be treated under continuous use of low-dose aspirin. Previous studies from Western countries have shown that proton pump inhibitors (PPI) are first-line drugs for the treatment of gastrointestinal injury associated with non-steroidal anti-inflammatory drugs (NSAIDs) including aspirin[9-11]. However, there are no prospective studies focusing on low-dose aspirin, nor are there studies from Japan where Helicobacter pylori (H. pylori) infection rate is high.

We conducted a prospective, multicenter controlled study to compare the therapeutic effects of ordinary dose of PPI and histamine 2 receptor antagonists (H2RA) on gastroduodenal ulcers during continuous use of low-dose aspirin.

Four patients in the PPI group and 4 patients in the H2RA group dropped out of the study because they refused to undergo endoscopy when their symptoms disappeared or they were moved to another hospital. Medication compliance rate was as high as 80% or more among patients excluding dropouts. Twenty-six patients in the PPI group and 26 in the H2RA group qualified for analysis.

Buffered aspirin tablets (Bufferin 81) and enteric coated tablets (Bayaspirin 100) were continuously used by 11 and 15 patients, respectively, in the PPI group and by 10 and 16 patients, respectively, in the H2RA group. Two patients in each group had used NSAIDs as needed for headache (diclofenac sodium in 3 patients and zaltoprofen in 1 patient). H2RA had been used before enrollment in 3 patients in the PPI group (usual dose of ranitidine, usual dose of nizatidine, and half dose of famotidine respectively) and in 4 patients in the H2RA group (usual dose of famotidine, usual dose of ranitidine, usual dose of nizatidine, and half dose of nizatidine respectively). No patients had used PPI before enrollment. If patients assigned to the H2RA group had a history of famotidine use, they were administered ranitidine 300 mg.

There were no significant differences in median age, sex, underlying disease, smoking status, H. pylori infection, prevalence of ulcers before treatment, or lesion site between the PPI group and the H2RA group (Table 1).

The therapeutic effects were endoscopically evaluated as healed in 23 of 26 patients in the PPI group and in 22 of 26 patients in the H2RA group, with no significant difference between the groups (Figure 1A).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Gastroduodenal ulcer healing was endoscopically achieved in proton pump inhibitors and histamine 2 receptor antagonists group patients. A: Gastroduodenal ulcer healing was endoscopically achieved in 88.5% of proton pump inhibitors (PPI) group patients and in 84.6% of histamine 2 receptor antagonists (H2RA) group patients. There was no significant difference between the two groups; B: Gastroduodenal ulcer healing in patients with non-pangastritis was endoscopically achieved in 88.2% of PPI group and in 86.7% of H2RA group. There was no significant difference between the two groups. NS: Not significant.

Three patients in the PPI group were evaluated as not healed, including 2 evaluated as reduced and 1 evaluated as unchanged. In the 2 patients evaluated as reduced, multiple ulcers were observed at the antrum of the stomach and ulcers 10 and 5 mm in maximum diameter were reduced to 3 and 2 mm, respectively, after treatment. In the patient evaluated as unchanged, the use of 100 mg aspirin enteric coated tablets was continued for the treatment of angina pectoris, endoscopy revealed a solitary ulcer 5 mm in diameter at the body of the stomach, and there was no evidence of H. pylori infection.

Four patients in the H2RA group were evaluated as not healed, including 3 evaluated as reduced and 1 evaluated as unchanged. In the 3 patients evaluated as reduced, solitary ulcers 15 mm at the antrum of the stomach, 15 mm at the body of the stomach, and 3 mm at the antrum of the stomach were all reduced to 2 mm after treatment. In the patient evaluated as unchanged, the use of 162 mg buffered aspirin tablets was continued for the treatment of old myocardial infarction, multiple ulcers were observed at the antrum of the stomach 3 mm in maximum diameter on endoscopy, there was no evidence of H. pylori infection, diclofenac sodium was prescribed as needed, and usual dose of ranitidine had been administered before assignment to treatment group.

Improvement of abdominal symptoms was evaluated after treatment by comparing the GSRS scores before and after treatment. Pretreatment scores indicated that abdominal symptoms were uncommon in both the PPI group and the H2RA group. The scores were not significantly reduced after treatment as compared with before treatment in either group (Figure 2). No side events were observed in either group.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Improvement of abdominal symptoms was evaluated using gastrointestinal symptom rating scale scores. The scores were not significantly reduced after treatment as compared with before treatment in both proton pump inhibitors (PPI) and histamine 2 receptor antagonists (H2RA) groups. ^aNS vs before. NS: Not significant.

For research about influence of acid secretion, both groups were subdivided into 2 groups based on whether the patient had pangastritis or not. There were 17 patients in the PPI group with pangastritis and 15 in the H2RA group. The therapeutic effects were not significantly different between the non-pangastritis groups (Figure 1B).

One of the major adverse events associated with low-dose aspirin is gastrointestinal bleeding. However, screening endoscopy is widely used in Japan and peptic lesions are frequently pointed out on screening endoscopy before the development of overt gastrointestinal bleeding. Therefore, the subjects of this study were those found to have gastroduodenal lesions on endoscopy without gastrointestinal bleeding as a complication.

Low-dose aspirin-induced mucosal defect is commonly 5 mm or less[3]. Considering that lesions 5 mm or less in size can cause bleeding, and if left untreated, may be enlarged, we believe that small ulcers should be treated. Thus, patients with ulcers 3 mm or more were included in this study.

Controlled studies of the therapeutic effects of PPI vs H2RA for NSAIDs-associated gastric ulcers have shown that the healing rate after 8 wk treatment was significantly higher in the PPI group than in the H2RA group[9-11].

A recent controlled study investigated the preventative effects of PPI vs H2RA on the occurrence of bleeding ulcers associated with low-dose aspirin in patients not infected with H. pylori and concluded that PPI are significantly more effective than H2RA in preventing the occurrence of bleeding ulcers and abdominal symptoms[13]. However, the results of their preventative effects against ulcers cannot be extrapolated to the healing of ulcers. On the other hand, it was reported that patients who take H2RA with low-dose aspirin had fewer peptic ulcers than patients who take placebo[14].

In the present study, the healing rate of gastroduodenal ulcers associated with low-dose aspirin was similar in the PPI group and the H2RA group. This may be explained by the facts that half of the subjects included in this study were infected with H. pylori, the subjects were limited to the Japanese, and they had a lower ability to secrete gastric acid. Previous studies involving patients with duodenal ulcers have shown that average maximum gastric acid secretion was 21.9 mEq/h for Japanese men and 43.2 mEq/mL for American men[15,16]. However, a retrospective study from Western countries investigated the effects of H2RA (famotidine 20-40 mg or ranitidine 150-300 mg) and PPI (omeprazole 20 mg) on lowering the risk of gastrointestinal bleeding in low-dose aspirin users and concluded that both drug classes have similar effects on preventing bleeding[17]. Further studies are needed to evaluate the therapeutic effects of both PPI and H2RA on low-dose aspirin-induced ulcers.

One patient evaluated as unchanged in the H2RA group developed an ulcer 3 mm in diameter, was not infected with H. pylori, and had a history of diclofenac sodium on an as-needed basis. One other patient in the PPI group had used diclofenac sodium as needed before assignment, but the ulcer was healed after 8 wk treatment. PPI are also reported to be effective when combined with low-dose aspirin and NSAIDs[18]. The combined use of low-dose aspirin and NSAIDs is known to be associated with an increased risk of gastrointestinal bleeding[3]. When low-dose aspirin is used alone, H2RA is expected to heal ulcers, but it may be better to choose PPI in high risk cases in which low-dose aspirin and NSAIDs are combined. Even in these cases, PPI are considered to be more effective than H2RA for 4 wk treatment[11,18].

The patient evaluated as unchanged in the PPI group had an ulcer 5 mm in size and was not infected with H. pylori. This case indicated that some ulcers are not healed even when treated with PPI. Such cases require further investigation for their appropriate treatment.

In this study, none of the patients were treated with combined aspirin and other antiplatelet drugs. However, recently use of combinations of low-dose aspirin and other antiplatelet drugs (e.g. clopidogrel and ticlopidine) have been increased, especially after coronary bypass graft surgery. PPI decrease clopidogrel’s inhibitory effect on platelets[19]. It may be better to choose H2RA when low-dose aspirin and clopidogrel are combined. On the other hand, clopidogrel is not associated with an increased risk of gastrointestinal bleeding when used alone, but is associated with a 7.7-fold increased risk when used in combination with low-dose aspirin[20]. Further studies are required to investigate the preventive and therapeutic effects of PPI and H2RA on gastrointestinal events in the presence of combined aspirin and other antiplatelet drugs.

In conclusion, the healing rate of gastroduodenal ulcers was greater than 80% after 8-wk treatment with PPI or H2RA during continuous use of low-dose aspirin, with no significant difference between the two groups.