Belching symptoms in gastroesophageal reux disease, and prevalence and clinical characteristics of belching in Japanese adults

Belching is the act of expelling gas from the stomach or esophagus noisily through the oral cavity. Although it is a physiological phenomenon, belching may also be a symptom of upper gastrointestinal diseases such as gastroesophageal reux disease (GERD). A detailed epidemiology of belching has not yet been reported. The aim of this study was to examine the prevalence and clinical characteristics of belching in Japanese adults. Methods We analyzed 1,998 subjects using data from a previous study of the association between GERD, psychological stress, and sleep disturbances in Japanese adults. Belching was evaluated according to the score to question 11 (‘Do you burp a lot?’) on the frequency scale for the symptoms of GERD (FSSG): 0 (never), 1 (occasionally), 2 (sometimes), 3 (often), and 4 (always). We also collected the clinical parameters, endoscopic ndings, and data according to the Athens Insomnia Scale (AIS), Rome III questionnaire, and Hospital Anxiety and Depression scale (HADS). The of in adults


Introduction
Belching is the act of expelling gas from the stomach or esophagus noisily through the oral cavity (1,2).
Generally, healthy adults experience belching several times a day as a physiological phenomenon.
Although belching is not a speci c symptom of upper gastrointestinal (GI) diseases, patients with gastroesophageal re ux disease (GERD) commonly complain of belching (3). Gastric belching is caused by transient lower esophageal sphincter relaxation, which is similar to the major pathogenic mechanism of GERD (4). Several studies have reported the prevalence and clinical characteristics of re ux symptoms in the general population (5,6). However, a detailed epidemiology of belching has not been elucidated.
The aim of this study was to examine the prevalence and clinical characteristics of belching using data from a previous study on the association between GERD, psychological stress, and sleep disturbances in Japanese adults (6).

Data
This study was performed using data from our previous study. A detailed methodology has been previously described (6). Data of 1,998 Japanese subjects with annual health check-ups at Kashiwara Municipal Hospital were analyzed. These included clinical parameters such as age, sex, smoking and alcohol drinking status, upper gastrointestinal endoscopy ndings, and the results of a self-report questionnaire comprising a frequency scale for the symptoms of GERD (FSSG) (7), Athens Insomnia Scale (AIS) (8), Rome III questionnaire (9), and Hospital Anxiety and Depression scale (HADS) (10).
Patients who continuously took acid-suppressing drugs, had active peptic ulcer disease, or had a history of upper GI surgery were excluded. This study was approved by the Ethics Committee of Kashiwara Municipal Hospital. Written informed consent was obtained from all subjects, and all procedures performed conformed to the Declaration of Helsinki.
De nition of GERD GERD is divided into re ux esophagitis and non-erosive re ux disease (NERD). Re ux esophagitis was de ned as the presence of an esophageal mucosal break, according to the Los Angeles classi cation (Grade A and more) (11). NERD was de ned as the absence of an esophageal mucosal break and an FSSG score ≥ 8 points (7). Subjects without a mucosal break and re ux symptoms (FSSG score < 8) were classi ed as subjects without GERD.

Statistical analysis
Values are expressed as mean ± SD and numbers (frequency) for continuous and categorical variables, respectively. Comparisons of categorical data between groups were performed using the chi-square test, while data from each group were statistically analyzed using the Kruskal-Wallis test. P-values < 0.05 were considered signi cant. A backward stepwise multiple logistic regression model was created to identify independent factors associated with belching. First, we analyzed several factors including age (< 75 or ≥ 75 years), sex (male or female), body mass index (BMI) calculated by body weight divided by the squared height (< 25 or ≥ 25 kg/m 2 ), smoking habits (current smoker or non-smoker), alcohol drinking habits (frequent drinker or infrequent/non-drinker), re ux esophagitis (present or absent), AIS scores (< 6 or ≥ 6, absence or presence of sleep disturbances) (8), functional dyspepsia (FD) (present or absent based on Rome IV criteria) (9), and HADS scores (< 8 or ≥ 8, absence or presence of anxiety/depression) (10). We calculated the odds ratio (OR) with 95% con dence intervals (CI) and excluded statistically insigni cant factors using the Wald test. All statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). More precisely, it is a modi ed version of R commander designed to add statistical functions frequently used in biostatistics (12).

De nition of signi cant belching
In our cohort, the prevalence of belching in GERD was 64.3% with an FSSG Q11 score ≥1 (occasionally or more frequent), 33.6% with scores ≥2 (sometimes or more frequent), 13.0% with scores ≥3 (often or more frequent), and 3.1% with a score of 4 (always). Since a previous outstanding study by Klauser showed that the prevalence of belching in patients with GERD was 40%-49%, we de ned signi cant belching in this study as subjects whose FSSG Q11 score was ≥2 (sometimes or more frequent).

Clinical characteristics of belching
We compared the clinical parameters and questionnaire scores of subjects with and without belching. There were no signi cant differences in age, BMI, and alcohol drinking or smoking habits between the two groups. However, subjects with belching were predominantly male and more commonly had FD but not re ux esophagitis. In addition, the HADS and AIS scores in subjects with belching were signi cantly higher compared to subjects without belching (Table 1). Table 2 shows an analysis of the logistic regression model. After adjustment of statistically factors by univariate analysis, male sex, presence of FD, anxiety/depression (HADS score ≥8), and sleep disturbances (AIS score ≥6) were signi cantly associated with belching.

Discussion
This is the rst epidemiological study of belching in Japanese adults. We found that the prevalence of belching was 17.3%. Male sex, FD, anxiety/depression, and sleep disturbances were associated with belching, while age, BMI, presence of re ux esophagitis, alcohol drinking, and smoking status were not.
Belching is divided into two distinct types, namely: gastric belching (GB) and supragastric belching (SGB) (1,2). GB involves swallowed air in the stomach triggering a transient lower esophageal sphincter relaxation. Gastric air that ows into the esophagus increases the pressure of the proximal portion, causing upper esophageal sphincter relaxation. This results in air out ow from the esophagus into the oral cavity (1,2). The mechanism of SGB differs in that contraction of the diaphragm induces a negative pressure in the esophagus. Air ows from the pharynx during relaxation of the upper esophageal sphincter and directly expels into the esophagus without entering the stomach (1, 2). However, it is di cult to distinguish these two types of belching without esophageal impedance pH monitoring (1,2,13). It is especially impossible in an epidemiological study. Therefore, we discuss identi ed factors associated with belching for both GB and SGB.
The reason for a male predominance in belching is still unknown. Saito et al. reported that men eat food faster than women (14), suggesting that men may swallow air more often during eating. In addition, men generally might prefer carbonated alcohol drinks and they do not hesitate to belch when compared to women. Although there is no sex difference in excessive SGB (1,2), the male predominance may be seen in GB.
Although it is not speci c, belching is one of the symptoms in patients with FD. Several studies demonstrated that 59-80% of patients with FD reported frequent belching (15,16). Conchilo et al. examined ten patients with FD and ten controls using esophageal impedance pH monitoring (17). They found that the incidence of air swallowing in patients with FD was signi cantly higher compared with controls (17). These ndings support our results.
The association between belching and anxiety/depression is uncertain. A high prevalence of anxiety disorders has been described in patients with excessive SGB. Among such patients, belching often increases during stressful events (2,18). These ndings might explain the positive association between anxiety/depression and the belching observed in this study.
Although belching rarely occurs during sleep, we found a signi cant association between belching and sleep disturbances. Similarly, a recent study by Hyun demonstrated that sleep disturbances were associated with belching (OR 1.59; 95% CI 1.24-2.03) in a cross-sectional study of 4,948 subjects (22). It might be related to the brain-gut axis which is involved in the pathogenesis of functional gastrointestinal disorders.
This study has some limitations. First, we assessed belching using only one question in the FSSG and de ned belching as a score ≥ 2 (sometimes, often, or always) (7). Although FSSG is speci c for GERD (7), there is currently no speci c questionnaire for belching. It is di cult to evaluate the number of daily belching in an epidemiological study. Second, we did not include other confounding factors that could affect belching. Factors such as intake of carbonated drinks and speed of food intake should be included in future studies.
In conclusion, to our knowledge, this is the rst report on the epidemiology of belching in Japanese adults. We also clari ed several clinical characteristics of belching. Since a recent study showed that excessive belching was associated with proton pump inhibitor refractory GERD (23), understanding the epidemiology of belching is important. Data are expressed as mean ± SD or number (frequency).