Endoscopy-diagnosed erosive esophagitis has been increasing in Asia over 2 decades, with a reported prevalence of 3.3 to 8.5% in Korea, China and Japan except Taiwan (ranging from 17.3 to 18.3%) which is similar to other Western developed countries.^1-6 The reasons for the high prevalence rate of erosive esophagitis are not clearly demonstrated, however, it could be related with western diet, increased obesity and aging of population.⁶ The prevalence rate of erosive esophagitis was 8%, in prospective nationwide multicentre studies of 25,536 health check-up Koreans.⁷ This prevalence rate was lower than that in Western countries.⁸ Association between high Helicobacter pylori infectivity and relatively low obesity rate has long been expected to maintain low prevalence of erosive esophagitis in Koreans.

The potential risk factors of erosive esophagitis are hiatal hernia, gender, hyperlipidemia, diabetes, alcohol, smoking, as well as H. pylori infection and obesity. Among them, H. pylori infection and obesity status are important on the prevalence of reflux esophagitis and have been studied throughout Asia.^9,10H. pylori seropositivity has been known to protect against gastroesophageal reflux in Asia, especially in the Far East.^11-13 Furthermore, patients with H. pylori infection have less severe gastroesophageal reflux disease,^14,15 however, H. pylori infection does not always influence gastroesophageal reflux disease, and the effect of eradication of H. pylori on reflux esophagitis is controversial.^16-20 Given the regional differences in seroprevalence of H. pylori in Korea,²¹ multicenter, prospective studies on the risk factors of reflux esophagitis seem to be important.

Recently there was a increase in obese patients as defined by those with body mass index (BMI) greater than 25 kg/m² in Koreans which are lower than that in the Western countries.²² Korea Centers for Disease Control and Prevention studied urban and rural patients with obesity and reported a signigicant differences, namely, 30.3% and 36.5%, respectively.²³ Thus, we need to know the change of seroprevalence of H. pylori and obesity influencing erosive esophagitis as well as other risk factors compared with previous Korean data. And also, by our data obtained from 8 health promotion centers, risk factor analysis of erosive esophagitis according to dwelling places is useful in elucidation of environmenal factors.

Therefore, we conducted this prospective multi-center study to investigate the interval change of prevalence of erosive esophagitis in healthy subjects and to find risk factors related to erosive esophagitis according to dwelling places.

Study Sample

A prospective study was carried out from January 2011 through December 2011 at 8 healthcare centers throughout Korea. Individuals with a history of gastrointestinal surgery or systemic diseases requiring chronic medication except hypertension and diabetes mellitus were excluded. This study was reviewed and approved by the Institutional review boards of the participating hospitals, and all participating subjects provided their written, informed consent. Residency status was divided into 2 types, according to big cities (Gyeonggi and Busan) and small and medium cities (Jeolla, Chungcheong, Gangwon and Jeju).

Questionnaire

Before endoscopy, clinical interviews based on a structured questionnaire were conducted, under the supervision of a well trained interviewer, to assess personal and clinical data of all subjects. The questionnaire included questions about demographic data, presence of upper gastrointestinal symptoms, comorbidities, history of H. pylori eradication, drug history including NSAIDs and antibiotics, alcohol consumption, smoking history, consumption of dairy products, and family history of gastric cancer.²¹ BMI, cholesterol, triglyceride (TG), glucose level and H. pylori infection status were recorded.

Endoscopic Findings

Endoscopic reflux esophagitis was diagnosed if definite erosions were present. Minimal change esophagitis was excluded because of weak interobserver agreement.²⁴ The severity of esophagitis was graded from A through D, according to the Los Angeles (LA) classification.²⁵ The presence of hiatal hernia was endoscopically diagnosed if the distance from the esophagogastric junction to the diaphragmatic impingement was more than 1 cm. The esophagogastric junction is defined as the proximal margin of the gastric mucosal fold. Edematous longitudinal fold of gastric mucosa taking a polypoid form was diagnosed as sentinel polyp. Benign peptic esophageal stricture was measured endoscopically and radiologically. Barrett's esophagus was suspected when endoscopic findings in the distal esophagus appeared to suggest gastric metaplasia, for example, salmon-pink colored columnar esophagus or presence of tongue-like projections.

Helicobacter pylori Serology Examination

Blood samples were collected from all participants before endoscopy. The samples were immediately centrifuged and stored at -70℃. Assessment of H. pylori infection was based on enzyme-linked immunosorbent assay (ELISA) for anti-H. pylori IgG using a Genedia kit (Genedia H. pylori ELISA; Green Cross Medical Science Corp, Eumsung, Korea) and duplicate determinations according to the manufacturer's guidelines. The Genedia kit makes use of an H. pylori antigen obtained from Korean H. pylori strains and has a sensitivity and specificity in Korean adults of 97.8% and 92.0%, respectively. The cut-off optical density (OD 450 nm) of H. pylori IgG was 0.406.

Statistical Methods

Continuous variables were analyzed by Student's t test. Categorical variables were analyzed using chi-square (χ²) test or Fisher's exact test. Univariate logistic regression and multivariate logistic regression analysis were used for the analysis of independent risk factors, which were expressed as the odds ratio (OR) and 95% confidence interval (CI). All P-values of less than 0.05 were considered statistically significant.

Characteristics of Participants

Of the 4,086 patients, 4,023 were eligible for the study. Sixty-three patients were excluded because of incomplete filling of a questionnaire. Mean age was 48.7 ± 11.3 years, and 2,358 (58.6%) patients were men. The clinical and demographic data of the patients are summarized in Table 1, including sex, age, H. pylori IgG, hiatal hernia, Barrett's esophagus, BMI, lipid profile, glucose level, history of H. pylori eradication, alcohol consumption, smoking history and place of residence. H. pylori seropositivity was detected in 2,407 (59.8%) patients. Seroprevalence of H. pylori was increased with older age (40-49 years, 55.5%; 50-59 years, 60.20%; above 60 years, 62.6%). There was no gender difference of H. pylori infection status.

Prevalence and Grade of Erosive Esophagitis

The prevalence rate of reflux esophagitis in this study was 8.8%. The distibution of erosive esophagitis was LA-A (6.4%), LA-B (2.1%), LA-C (0.3%) and LA-D (0.1%). One mucosal break was more common than multiple. The degree of mucosal breaks (< 5, 5-10 and 10-20 mm) was particulary pronounced in 5 mm in all age groups (≤ 39 years, 73.3% vs. 26.7% vs. 0.0%; 40-59 years, 72.7% vs. 21.8% vs. 3.2%; ≥ 60 years, 70% vs. 26% vs. 2%). Absence of confluent erosions were most common in all age groups. The prevalence of erosive esophagits was higher in men than in women, especially with LA-A, LA-B and LA-C. The prevelance rate of erosive esophagitis was similar in each aging group (≤ 39 years, 8.8%; 40-59 years, 9.1%; ≥ 60 years, 7.9%) (Table 2). The prevalence of seropositivity in residents of big cities (Gyeonggi and Busan) and in small and medium cities (Jeolla, Chungcheong, Gangwon and Jeju) was 56.6% and 60.5%, respectively, while the prevalence rates of reflux esophagitis was 7.1% and 9.4%, respectively (Fig. 1). Endoscopically suspected Barret esophagus was 0.8% (male, 0.9%; female, 0.5%). Stricture was not detected and sentinel polyp was very rare (0.1%).

Risk Factors for Reflux Esophagitis

Hiatal heria was 2.6% and had no gender difference. Small size of hiatal hernia was more common (1 cm < size ≤ 2 cm, 54.5%). Total frequency of BMI higher than 25 kg/m² was 35.8%. The frequency of obesity were significantly different between big, and small and medium-sized cities (26.9% and 38.8%, respectively; P < 0.001) (Fig. 2).

Univariate analysis identified significant differences regarding sex, H. pylori seronegativity, hiatal hernia, Barrett's esophagus, BMI ≥ 25 kg/m², TG ≥ 125 mg/dL, fasting glucose level ≥ 126 mg/dL, history of H. pylori eradication, alcohol consumption, smoking habit, residence in big cities and salty diet (Table 3). Multivariate analysis revealed that several factors were independently associated with an increased risk of reflux esophagitis: H. pylori seronegativity (OR, 1.91; 95% CI, 1.48-2.46), BMI ≥ 25 kg/m² (OR, 1.49; 95% CI, 1.08-2.06), TG ≥ 150 mg/dL (OR, 1.65; 95% CI, 1.08-1.06), fasting glucose level ≥ 126 mg/dL (OR, 1.72; 95% CI, 1.06-2.81) and hiatal hernia (OR, 3.10; 95% CI, 1.86-5.16). The OR of residents in a big city for the prevalence of reflux esophagitis was 0.59 (95% CI, 0.39-0.90) (Table 4).

The prevalence of reflux esophagitis in Korea has increased since the 1990s. In 1999, the prevalence of reflux esophagitis in patients with gastrointestinal symptoms was 5.3%.²⁶ The prevalence of erosive esophagitis in patients undergoing regular health check-up was 8.0% in 2006.⁷ In this study, the prevalence of reflux esophagitis was 8.8%, which is similar to the previous study.

Seroprevalence of H. pylori were 59.5% and 59.8% in 2005 and 2011, respectively. The relationship between H. pylori and reflux esophagitis is not clear yet. In a recent large case-control study, H. pylori seropositivity was inversely associated with the risk of reflux esophagitis, and the prevalence of H. pylori infection was lower in reflux esophagitis cases than in controls (38.4% vs. 58.2%, P < 0.001).¹³ In our large sample, the prevalence of H. pylori infection was also lower in reflux esophagitis cases than in patients without reflux esophagitis (44.6% vs. 61.3%, P < 0.0001) and the OR of H. pylori seronegativity for the development of reflux esophagitis was 1.91 (95% CI, 1.48-2.46). H. pylori infection was inversely correlated with the risk of erosive esophagitis.

The association between H. pylori seropositivity and residence location is well known.²¹ According to geographic area in 2005, seroprevalence was 57.6% and 60.5% in big cities (Seoul and Gyeonggi, respectively), while higher than 60% in small and medium cities. The seroprevalence of H. pylori between 2005 and this study was not changed in big and small-medium cities. Differences in socioeconomic condition, sanitation and hygienic practices may explain this result. Taken together, the seroprevalence was slightly lower in big city residence than small and medium sized cities, suggesting that this difference could not explain the different prevalence rate of erosive esophagitis according to dwelling status.

Asides from H. pylori infection, obesity is an another major risk factor for erosive esophagitis.^27,28 However, recent studies on the association between BMI and erosive esophagitis have resulted in heterogenous data.^29-32 There are no large-scale prospective studies involving healthy subjects with erosive esophagitis in the aspect of dwelling places. Relationship between obesity and erosive esophagitis is well known,²⁸ however, the gender effect is controversial.^33-35 Our study showed that rate of obesity in big city dwelling was less frequent than that in small and medium cities after adjusting for potential confounding factots. According to KNHANES obseity data from 1998 to 2009 (1998, 2005 and 2007-2009), Korean with a BMI ≥ 25 kg/m² was found to be 25.9%, 34.8% and 35.8% in men, 27.5%, 28.3% and 26.1% in women, respectively.²² This result suggests that the prevalence of obesity was recently stagnated, and that there is a decreasing tendency in women. Our study shows that the prevalence rate of erosive esophagitis was lower in big city dwelling than in small and medium cities dwelling (7.1% and 9.4%, respectively) due to less frequent rate of obesity (P < 0.001) and these data were obtained after adjusting for confounding factors, including H. pylori infection, TG, glucose and hiatal hernia. This lower rate of BMI ≥ 25 kg/m² in big city residence may be explained by the fact that they may have a chance to receive a regular private health screening and engage in regular exercise maintaining a healthy weight.

Many other studies have found that hiatal hernia is associated with a high risk of reflux esophagitis.^36-39 In our study, according to univariate analysis hiatal hernia (OR, 3.02; 95% CI, 1.87-5.16) was significantly associated with reflux esophagitis (all P < 0.0001).

Factors that are consistently reported to be associated with reflux esophagitis include advanced age, diabetes, hyperlipidemia, alcohol and smoking.^19,40-45 In the present study, multivariate analysis identified high TG (OR, 1.65; 95% CI, 1.08-2.07) and elevated fasting glucose (OR, 1.73; 95% CI, 1.06-2.81) level as risk factors. Reports of the effects of smoking and alcohol consumption on the development of reflux esophagitis have been inconsistent.^44,45 In our study, according to univariate analysis alcohol consumption (OR, 2.19; 95% CI 1.74-2.76) and smoking habit (OR, 1.66; 95% CI, 1.28-2.16) were significantly associated with reflux esophagitis (all P < 0.0001).

The present study has several potential limitations. First, 527 (13.1%) patients with H. pylori eradication were included. The reasons for eradication were peptic ulcers (24.9%), gastroduodenitis (33.7%), patients' choice (35.9%) and family history of gastric cancer (2.6%). However, these subjects were equally distributed among cases and controls and there was no statistically significant difference between patients with and without reflux esophagitis, with respect to H. pylori eradication. Second, H. pylori infection status was evaluated only by serologic testing. Therefore, we did not assess the current or past infection status. Third, analysis of risk factors did not take into account the severity of reflux esophagitis.

In conclusion, the prevalence rate of erosive esophagitis and its risk factors in this study were similar to nationwide study in 2006. H. pylori seropositivity and residence in a large city were inversely related to the risk of reflux esophagitis in a large-sample, multicenter study of the Korean population. The degree and duration of obesity would be important risk factor for the development of erosive esophagitis in the future. Factors associated with metabolic syndrome including high TG and elevated fasting glucose level and hiatal hernia were also related to the risk of reflux esophagitis.

This study is meaningful in that as it has provided clues regarding the prevalence of erosive esophagitis and risk factors in healthy subjects. Further studies are necessary to determine the potential risk factors for the development of erosive esophagitis and we should pay attention to the changes in the prevalence of obesity in Koreans.