Introduction

Since Helicobacter pylori (H. pylori) was discovered by Warren and Marshall in 1982 [1], it has been found to be related to various gastroduodenal diseases, including peptic ulcer [2], gastric cancer [3] and mucosa-associated lymphoid tissue (MALT) lymphoma [4]. Eradication of H. pylori inhibits the recurrence of peptic ulcer [5], ameliorates MALT lymphoma [6] and lowers the risk of gastric cancer [7]. About 50% of people worldwide including Japan are thought to be infected with this bacillus [8], and prevalence increases with age in Japan, particularly in those over 40 years [9].

Although some reports have shown that H. pylori-positive patients tend to have dyspepsia [10, 11], the relationship between H. pylori infection and dyspepsia is controversial. The prevalence of H. pylori in Japan has been decreasing with steady improvements in sanitary conditions. Moreover, some reports have shown using endoscopy that H. pylori-negative patients do not have structural abnormalities [10]. The 2005 American College of Gastroenterology (ACG) guidelines for the management of dyspepsia [12] recommend tests for H. pylori infection among dyspeptic patients without alarm features as well as among H. pylori-positive patients, and proton pump inhibitors for H. pylori-negative patients.

It is important to recognize the prevalence of H. pylori in dyspeptic patients from the standpoint of eradication cost in Japan where prevalence of H. pylori is high [9]. And endoscopic findings are important for the indications that early endoscopy can provide. Japanese guidelines for dyspepsia can only be established after the prevalence of H. pylori infection and endoscopic findings in dyspeptic patients have been clarified. The present study evaluates H. pylori status and endoscopic findings among outpatients with dyspepsia at a Japanese hospital.

Materials and methods

Subjects

We performed a prospective study of 4,331 new outpatients (1,914 men and 2,417 women; average age, 53.4; 53.1 and 53.7 years for men and women, respectively) at the Department of General Medicine in Oita University Hospital between 2002 and 2006.

Each patient completed a questionnaire regarding their reasons for attending the hospital, containing questions such as “What is wrong with you today?” Then a general physician obtained a detailed medical history. Patients whose chief complaint was dyspepsia were assigned to the dyspeptic group. Dyspepsia was regarded as “pain or discomfort centered in the upper abdomen”. Symptoms associated with or described as discomfort were upper abdominal fullness, early satiety, bloating, belching, or nausea. Patients with heartburn or diarrhea were not considered dyspeptic. Patients with a history of H. pylori eradication were excluded from the study. The major chief complaints among the new outpatients were dyspepsia (n = 258), headache (n = 243), fever (n = 217), cough (n = 200), chest pain (n = 178), back pain (n = 160). Use of non-steroidal anti-inflammatory drugs (NSAIDs) and anti-ulcer agents as well as smoking status were also recorded.

We categorized the patients into groups according to age as young (age ≤39 years); middle-aged (age 40–69 years) and elderly (age ≥70 years).

All subjects provided written, informed consent and the Ethics Committee of Oita University the approved study.

Evaluation of H. pylori status

We evaluated H. pylori status using a rapid urine test (RAPIRUN® H. pylori antibody, Otsuka Pharmaceutical Co., Tokyo, Japan). The reported sensitivity, specificity, positive and negative predictive values of the kit are 95.3, 96.7, 95.3 and 96.7%, respectively [13]. Antibodies in urine from each patient were immediately measured after collection, using the kit according the manufacturer’s instructions. The same skilled technician analyzed all urine samples using the kits and interpreted the results after 15–30 min.

Analysis of upper gastrointestinal endoscopic findings

We examined the endoscopic findings of 138 among 258 dyspeptic patients. A skilled gastroenterologist who was blinded to the H. pylori status of the patients but not to their questionnaire answers performed all endoscopic procedures and diagnosed the findings. Two endoscopists who were unaware of the H. pylori status of each patient independently conducted a retrospective endoscopic diagnosis. Details of gastric cancer, peptic ulcer or reflux esophagitis were recorded. Grade M reflux esophagitis was not regarded as reflux esophagitis. Esophageal hiatal hernia or endoscopic gastritis (atrophic gastritis, erythematous gastritis and erosive gastritis) was also recorded, but not regarded as significant endoscopic findings. When two or more endoscopic findings were found in the same subject, all findings were recorded.

Statistics

We performed all statistical analyses using the SPSS statistical package for Windows. Data were statistically analyzed using Student’s t test and the Chi-squared test. p values of <0.05 were considered significant.

Results

Analysis on the links between H. pylori and dyspepsia

The prevalence of H. pylori in 258 dyspeptic patients (107 males and 151 females) was 44.5% (115/258) and this increased with age (Fig. 1): 11.1% (2/18) at 10–19 years, 20.8% (10/48) at 20–29 years, 38.8% (14/36) at 40–49 years, and over 50% at age 50–79. Infection of H. pylori was less prevalent among those aged >80 than that at 70–79 years of age. Patients were divided according to age into young, middle-aged and elderly groups. The prevalence of H. pylori in each group tended to be lower among dyspeptic male than female patients (12.9 vs. 22.7% at young group and 45.0 vs. 58.4% at middle-aged group, and 68.0 vs. 90.0% at elderly group; p = 0.700, 0.150 and 0.070, respectively) (Table 1). Twenty-four in 258 (9.3%) dyspeptic patients used NSAIDs. NSAIDs use in males and females was equal (8.4% [9/107] vs. 9.9% [15/151]; p = 0.67). There was also no difference of NSAIDs use between H. pylori-positive and -negative patients (11.3% [13/115] vs. 7.6% [11/143]; p = 0.32).

Fig. 1
figure 1

Prevalence of H. pylori in 258 dyspeptic patients. Prevalence of H. pylori determined using rapid urine was 44.5% overall, and increased with age

Full size image
Table 1 H. pylori prevalence in dyspeptic patients by age and gender
Full size table

Endoscopic findings and relationship to H. pylori positivity

Endoscopic findings based on H. pylori status were evaluated in 138 patients who underwent endoscopy (Table 2). Overall, 18.1% (25/138) of these patients had peptic ulcers, 6.5% (9/138) had reflux esophagitis and 2.1% (3/138) had gastric cancer, which was also identified in 3.2% (3/138) of H. pylori-positive patients. There was no gastric cancer in H. pylori-negative patients. The prevalence of peptic ulcers was significantly higher among H. pylori-positive than negative patients (25.0% [23/92] vs. 4.3% [2/46]; p = 0.0029). All H. pylori-negative patients were in the scar phase and half of the H. pylori-negative patients with peptic ulcers patients were taking NSAIDs. The prevalence of H. pylori in patients with peptic ulcers was 92.0% (23/25). The prevalence of reflux esophagitis tended to be lower in H. pylori-positive than -negative patients (4.3% [4/92] vs. 10.8% [5/46]; p = 0.14). Of the 46 H. pylori-negative and 92 H. pylori-positive patients, 39 (84.7%) and 62 (67.3%), respectively, were considered endoscopically normal (no significant findings; p = 0.029). Among these endoscopically normal H. pylori-positive and -negative patients, use of NSAIDs and anti-ulcer agents did not significantly differ, whereas more H. pylori-negative patients smoked (p = 0.049).

Table 2 Endoscopic findings for dyspeptic patients
Full size table

Discussion

Dyspepsia is a complaint commonly seen in clinical practice [14]. Some reports [10, 11] have observed that dyspeptic symptoms are more frequent among H. pylori-positive patients. Although changes in gastric acid secretion by H. pylori infection or functional abnormalities in gastric movement might contribute to dyspepsia, the relationship has not been fully elucidated [15–17]. Shimatani et al. [10] reported that dyspepsia symptoms are more frequent among H. pylori-positive patients in Japan. Our results contradict these findings and might be due to differences in subjects (they studied patients undergoing medical check-ups). However, Sasaki et al. [18] and Kawamura et al. [19] found no relationship between dyspepsia and H. pylori status. These studies were also conducted among persons attending medical check-ups. Few reports have described the prevalence of H pylori with dyspepsia in a hospital setting. In our study, over half of the dyspeptic Japanese patients examined were negative for H. pylori. Therefore, factors other than H. pylori might play a role in the development of dyspepsia in the hospital setting. The prevalence of H. pylori in patients with dyspepsia tended to be lower in males than in females. Reports indicate that obesity and smoking are more important than H. pylori infection for dyspepsia [20, 21]. It is possible that many dyspeptic male patients are obese or smoke. More male than female patients smoked in the present study (data not shown), but we did not have sufficient information about obesity to draw any conclusions. These issues require further study.

The endoscopic findings of dyspeptic patients revealed peptic ulcers, reflux esophagitis and gastric cancer in 18.1, 6.5 and 2.1% of patients, respectively. An ACG technical review has shown that the incidences of these three conditions are 5–15, 5–15 and <2%, respectively [22]. Our results are consistent with these values.

Half of the H. pylori-negative patients with peptic ulcers used NSAIDs, which should be considered in a diagnosis of peptic ulcers in H. pylori-negative dyspeptic patients. Thomson et al. [23] evaluated endoscopic findings from dyspeptic patients, and found that peptic ulcers were more prevalent among patients using NSAIDs. Reflux esophagitis was also more frequent in H. pylori-negative patients [23]. Acid output might be higher among H. pylori negative patients [24]. No structural abnormalities on endoscopy were identified in 67.3 and 84.7% of H. pylori-positive and-negative patients and the difference was significant (p = 0.029). Shimatani et al. [10] examined young patients undergoing medical check-ups, and found no abnormalities on endoscopy in H. pylori-negative patients. Our results agree with their findings. However, the possibility of selection bias cannot be denied because the prevalence of H. pylori in patients who underwent endoscopy was higher than that among overall dyspeptic patients. Attending physicians might routinely order endoscopy for H. pylori-positive patients. Furthermore, we did not examine alarm features. The 2005 ACG guidelines for the management of dyspepsia [12] recommend endoscopy for dyspeptic patients with alarm features.

In conclusion, we investigated the prevalence of H. pylori and endoscopic findings in dyspeptic Japanese patients and found that over half of them were H. pylori-negative. Patients who are H. pylori-positive should undergo endoscopy to rule out gastric malignancy and peptic ulcer diseases.