Belching and gastroesophageal reflux disease (GERD) negatively impact the quality of life.^1-4 Previous studies in referral centers showed that almost half of patients with GERD had excessive belching as their main disturbing symptom, and supragastric belching (SGB) was demonstrated as the primary mechanism of belching in these patients.^5-8 An association between GERD and SGB has been suggested. SGB occurred more frequently in patients with GERD than in healthy controls, and SGB usually occurred in a close temporal association with reflux episodes.⁶ Moreover, behavioral interventions for SGB significantly reduced the number of SGB episodes and daytime esophageal acid exposure, and alleviated the proton-pump inhibitor (PPI)-refractory GERD symptoms.^2,5

SGB is voluntary, as attention and distraction affect the SGBs frequency.⁹ SGB initially occurred in response to the stress or unpleasant feelings in the throat, esophagus, chest, and epigastrium. Then, it subsequently changes into subconscious behavior without preceding gastrointestinal (GI) symptoms.^6,10-12 On the other hand, SGBs could elicit reflux episodes or induce esophageal symptoms by creating negative intrathoracic pressure with deep breathing or sucking air into the esophagus or contracting pharyngeal muscles to push the air into the esophagus.^8,13,14 Whether reflux episodes induced SGBs or SGBs induced reflux episodes is not clearly known. Twenty-four hour esophageal pH-impedance monitoring can demonstrate the temporal relationship between reflux episodes and SGBs, which will elucidate the interactions between reflux episodes and SGBs in patients with GERD and excessive belching. Thus, this study aims to evaluate gastroesophageal reflux characteristics and explore the interaction between reflux episodes and SGBs in patients with excessive belching with GERD symptoms by using 24-hour esophageal pH-impedance monitoring. Through this study, we hope to gain insight into the pathogenesis and potential treatment of the SGB-associated GERD subgroup, which is a significant group of patients with GERD.

Participants

We reviewed the medical records of consecutive adult patients who were referred for 24-hour esophageal pH monitoring at the Center of Excellence in Neurogastroenterology and Motility, Chulalongkorn University, Bangkok, Thailand. Patients who had typical gastroesophageal reflux symptoms with bothersome belching that met the Rome IV diagnostic criteria for the belching disorder (bothersome belching > 3 days a week for the last 3 months with symptom onset ≥ 6 months before diagnosis) were included.¹⁵ We excluded patients with a history of upper GI tract surgery, pregnancy, upper GI endoscopy 1 year before enrollment revealed severe reflux esophagitis (Los Angeles grades C and D), peptic stricture, and histology-proven Barrett’s esophagus. All participants completed the validated upper and lower GI symptoms questionnaires to evaluate the frequency and severity of their symptoms during 3 months before esophageal pH monitoring. The visual analog scale (VAS, 0-10) was used to assess the severity of the global GI and belching symptoms. The hospital anxiety depression symptoms questionnaires (HADS) were also completed. All patients underwent high-resolution esophageal manometry (HREM) and 24-hour esophageal pH-impedance monitoring during off PPIs and prokinetics for at least 7 days.

This study was conducted following the guidelines of the Declaration of Helsinki. All participants have given their written informed consent. The study protocol was reviewed and approved by the Institutional Review Board of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand (IRB No. 662/61).

Twenty-four Hour Esophageal pH-impedance Monitoring and Analysis

A 6-channel impedance catheter coupled with one pH sensor (Omega, MMS, Enschede, The Netherlands) was inserted transnasally and fixed the pH sensor at 5 cm above the localized upper border of the lower esophageal sphincter (LES), using HREM (Medtronic Inc, Shoreview, MN, USA). For impedance measurement, the sampling rate was 50 Hz, and the catheter provided recordings from 6 sensors pairs, each pair lying 2 cm apart (2-4, 4-6, 6-8, 8-10, 14-16, and 16-18 cm above the upper border of LES).¹⁶ The 24-hour esophageal pH-impedance monitoring was manually analyzed by 1 investigator with less than 30 seconds per analytic frame. “Pathological SGB” was defined as a repetitive rapid increase in impedance (≥ 1000 Ω) progressing in an aboral direction and not reaching the most distal esophageal sensor, followed by an oral direction returning for more than 13 times/day.^12,13 SGB episodes occurring less than 5-second apart were considered a single episode. We included only the patients with pathological SGB demonstrated by a 24-hour esophageal pH-impedance monitoring for further reflux and SGBs characteristics analysis. The reflux component was defined according to the international working group consensus.¹⁷ Liquid reflux was characterized as a retrograde 50% drop in impedance from baseline lasting for ≥ 4 seconds in the 2 distal impedance sensors. Mixed liquid-gas reflux was defined as gas reflux occurring during or immediately before liquid reflux. Gas reflux was defined as a simultaneous increase in impedance > 3000 Ω in any 2 consecutive impedance sites, with 1 site having an absolute value > 7000 Ω without swallows. Pure gas reflux events were not taken into account. Liquid and mixed reflux episodes were classified as acid and non-acid when reflux events were associated with a fall of esophageal pH < 4 and ≥ 4, respectively. Significant distal total esophageal acid exposure or positive pH monitoring (pH+) was defined using the Lyon Consensus as % time pH < 4 at distal esophagus > 6%, or between 4-6% plus any abnormal adjunctive parameters from impedance metrics or HREM.¹⁸

We classified reflux events according to the temporal association between reflux episodes and SGBs into (1) reflux episodes preceded by SGBs: SGBs occurred immediately (1 second) before reflux onset, (2) reflux episodes followed by SGBs: SGBs occurred within 10 seconds after a reflux onset,⁶ and (3) lone reflux: reflux episodes without SGBs occurred during 1 second before and 10 seconds after reflux onset (Fig. 1). Demographic and reflux characteristics were compared between patients with positive (pH+) and negative (pH–) tests.

Figure 1. We classified reflux events according to the temporal association between reflux episodes and supragastric belches (SGBs) into (1) reflux episodes preceded by SGBs: SGBs occurred immediately (1 second) before reflux onset (A), (2) reflux episodes followed by SGBs: SGBs occurred within 10 seconds after a reflux onset (B), and (3) lone reflux: reflux episodes without SGBs occurred during 1 second before and 10 seconds after reflux onset (C).

Statistical Methods

Data are expressed as mean ± standard deviation or median (interquartile range). Based on the data distribution pattern, the Mann-Whitney U test or the Student’s t test was used for continuous variables comparisons, and Fisher’s exact test or chi-square test was used for the categorical variables comparisons. The relationships between variables were expressed with Spearman’s correlation analysis. Statistical analyses were conducted using SPSS, version 28.0 (IBM, Chicago, USA), and a P-value < 0.05 is considered statistical significance.

We enrolled 74 patients who had excessive belching and symptoms suspected of GERD. Of these, 46 patients (62.2%) demonstrated a pathological SGB during 24-hour esophageal pH-impedance monitoring and were included in the analysis.

Baseline Characteristics of Patients With Pathological Supragastric Belch

The average patient’s age was 47.0 ± 13.0 years (range 17-71 years), and 34 patients (74.0%) were female. The mean body mass index (BMI) was 23.9 ± 4.2 kg/m². The average global GI symptoms severity score (VAS, 0-10) was 7.4 ± 1.6 with a belching severity score (VAS, 0-10) of 7.5 ± 1.7. The median duration of excessive belching was 6.5 months (range 6-12 months). Twenty-three of 46 patients (50.0%) had excessive belching, while 13 patients (28.3%) reported heartburn or regurgitation as their most disturbing symptom. Thirty-eight patients (80.4%) experienced an inadequate response to once or twice daily PPIs for at least 8 weeks. Twenty-one patients had one or more overlapping functional GI disorders diagnosed by Rome IV criteria, including functional dyspepsia (n = 21, 45.7%), functional constipation (n = 10, 21.7%), and irritable bowel syndrome (n = 8, 17.4%). Nine patients (19.6%) had depression and/or anxiety, as stated by HADS. HREM was performed only in the upright position and demonstrated hiatal hernia in twelve patients (26.1%). Five patients (11.0%) had major esophageal motility disorders (esophagogastric junction outflow obstruction (n = 2), absent contractility (n = 1), hypercontractile esophagus (n = 1), and distal esophageal spasm (n = 1) and 23 patients (50.0%) had ineffective esophageal motility diagnosed by the Chicago classification version 3.0.¹⁹

The mean duration of the pH-impedance study (meal excluded) was 22.2 ± 0.1 hours. Fifteen patients (32.6%) had positive pH monitoring with a median % time pH < 4 at distal esophagus (total acid exposure time [AET]) of 7.8% (5.4-9.2%). Age, BMI, the prevalence of hiatal hernia and major esophageal motility disorders, duration of the monitoring, and mean resting LES pressure were similar between patients with positive and negative pH monitoring. Gastroesophageal reflux characteristics of patients with positive and negative pH monitoring are shown in Table 1.

Table 1 . Clinical and Gastroesophageal Reflux Characteristics Compared Between Patients With Positive and Negative pH Monitoring

Characteristics	pH+ (n = 15)	pH– (n = 31)	P-value
Age (yr)	52.4 ± 12.1	45.0 ± 12.1	0.910
Body mass index (kg/m2)	24.2 ± 3.3	23.7 ± 4.5	0.743
Hiatal hernia	4 (26.7)	8 (25.8)	0.952
Mean resting LESP (mmHg)	22.6 ± 14.2	25.0 ± 14.0	0.600
Major esophageal motility disorders	1 (6.7)	4 (12.9)	0.705
Ineffective esophageal motility	11 (73.3)	12 (38.7)	0.067
Recording duration, hours	22.0 ± 0.8	22.4 ± 1.0	0.161
% time pH < 4 at the distal esophagus	7.8 (5.4-9.2)	0.8 (0.2-1.5)	< 0.05
GER episodes (times/day)	44.1 ± 19.7	26.7 ± 14.6	< 0.05
% acid reflux	41.8 ± 24.4	25.8 ± 16.6	< 0.05
Proximal extension (cm) from LES	15.0 (9.0-15.0)	9.0 (9.0-15.0)	0.136
Bolus contact time (sec)	14.4 (12.2-19.8)	12.1 (9.5-15.3)	0.083

pH+, patients with positive pH monitoring; pH–, patients with negative pH monitoring; LESP, lower esophageal sphincter pressure; GER, gastroesophageal reflux; LES, lower esophageal sphincter.

Data are expressed as mean ± SD, n (%), or median (interquartile range).

Supragastric Belches Characteristics and Association Between Supragastric Belches and Gastroesophageal Reflux Episodes

The average number of SGBs during the study period was 157 (113-242) episodes. 94.5 ± 6.2% of SGB episodes occurred in the upright position, and 54.2 ± 27.2% occurred during the 2-hour postprandial period. The longest repetitive SGBs duration was 2.0 (1.4-3.0) minutes, with the maximum SGBs rate of 14.6 ± 5.5 times/minute (range 5.0-25.0). Nearly half of the reflux episodes had SGBs before reflux onset (reflux episodes preceded by SGBs: 48.1 ± 21.0%), followed by lone reflux events (39.6 ± 19.0%), and reflux episodes with SGBs following (12.3 ± 10.2%).

Patients with pH+ significantly had a higher SGBs rate (205 [160-284] vs 132 [101-132] times/day, P < 0.05) and a higher reflux episodes preceded by SGBs (23 [10-33] vs 11 [6-16] times/day, P < 0.05) compared to patients with pH–. The proportion of SGBs immediately before reflux onset to total SGB episodes per day was not significantly different between patients with pH+ and pH– (0.11 [0.07-0.16] vs 0.07 [0.04-0.12], P = 0.150) (Fig. 2A). The difference in the number of reflux episodes between patients with pH+ and pH– was contributed mainly by reflux episodes preceded by SGBs (Fig. 2B). There was a significant positive correlation between the number of SGBs during the study period and % AET at the distal esophagus (r = 0.41, P < 0.05; Fig. 3A) and between the number of SGBs and the number of reflux episodes preceded by SGBs during the study period (r = 0.43, P < 0.05; Fig. 3B). In contrast, the proportion of the number of SGBs preceded by reflux episodes/total SGB episodes was markedly low in the whole patient group (0.02 [0.003-0.05]) and patients with positive and negative pH monitoring (0.03 [0.008-0.05] and 0.01 [0.001-0.04], respectively) (Fig. 4).

Figure 2. Numbers of supragastric belches (SGBs) classified as SGBs unrelated to reflux, SGBs within 10 seconds after reflux onset, and SGBs immediately before reflux onset during the study period in the pH-positive (pH+) and pH-negative (pH–) subgroups. The proportion of SGBs subtypes was not significantly different between pH+ and pH– (A), *P < 0.05 total numbers of SGBs during the study period in patients with pH+ vs pH–. Numbers of gastroesophageal reflux episodes classified as reflux episodes preceded by SGBs and reflux episodes without preceding SGBs during the study period in the pH+ and pH– subgroups (B). *P < 0.05 total numbers of reflux episodes during the study period and reflux episodes preceded by SGBs during the study period in patients with pH+ vs pH–.

Figure 3. Correlation analysis between the number of supragastric belches (SGBs) during the study period and % acid exposure time at distal esophagus (A) and between the number of SGBs and reflux episodes preceded by SGBs during the study period (B).

Figure 4. The proportion of supragastric belches (SGBs) preceded by reflux episodes to total SGB episodes during the study period comparing between the patients with positive (pH+) and negative (pH–) esophageal pH monitoring.

The prevalence of hiatal hernia and resting LES pressure were not significantly different between patients with pH+ and pH– (hiatal hernia 26.7% vs 25.8%, P = 0.952; LES pressure 22.6 ± 14.2 vs 25.0 ± 14.0 mmHg, P = 0.600). There was a higher prevalence of ineffective esophageal motility in patients with pH+ than pH– but it did not reach the statistical significance (73.3% vs 38.7%, P = 0.067). Patients with pH+ had longer acid contact time at the distal esophagus than patients with pH– (46.9 [21.5-75.0] seconds vs 13.3 (8.1-41.5) seconds, P < 0.05]. The bolus contact time (BCT) detected by impedance sensors was not significantly different between patients with pH+ and patients with pH– (14.4 [12.2-19.8] seconds vs 12.1 [9.5-15.3] seconds, P = 0.083).

Characteristics of Reflux Episodes Preceded by Supragastric Belches, Reflux Episodes Followed by Supragastric Belches, and Lone Reflux Episodes

Reflux episodes preceded by SGBs and reflux episodes followed by SGBs extended significantly more proximal, had longer acid contact time, and had longer BCT than lone reflux episodes (all P < 0.05). In addition, reflux episodes followed by SGBs had significantly more proportion of acid reflux than reflux episodes preceded by SGBs and lone reflux (both P < 0.05) (Table 2).

Table 2 . Characteristics of Reflux Episodes Preceded by Supragastric Belches, Reflux Episodes Followed by Supragastric Belches, and Lone Reflux Episodes

Characteristics	Reflux episodes preceded by SGBs	Reflux episodes followed by SGBs	Lone reflux episodes
Proximal extension (cm) from LES	9.0 (9.0-15.0)a	9 (9.0-16.5)b	9 (7.0-9.0)a,b
Acid contact time (sec)	24.7 (15.3-75.2)a	29.5 (17.3-43.8)b	15.9 (7.4-33.0)a,b
Bolus contact time (sec)	14.7 (10.0-19.4)a	14.7 (10.8-20.5)b	11.9 (9.7-14.3)a,b
% acid reflux	64.1 (50.0-90.1)c	100.0 (50.0-100.0)b,c	63.6 (41.7-91.5)b

^aP < 0.05 reflux episodes preceded by supragastric belches (SGBs) vs lone reflux episodes.

^bP < 0.05 reflux episodes followed by SGBs vs lone reflux episodes.

^cP < 0.05 reflux episodes preceded by SGBs vs reflux episodes followed by SGBs.

LES, lower esophageal sphincter.

Data are expressed as median (interquartile range).

This study demonstrated the characteristics of gastroesophageal reflux episodes and the interaction between reflux episodes and SGBs in patients with belching disorder according to the Rome IV diagnostic criteria using a 24-hour esophageal pH-impedance study. Nearly one-third (32.6%) of the patients with pathological SGB had positive pH monitoring. Around half of the reflux episodes in these patients were associated with SGBs, particularly the reflux episodes preceded by the SGBs. In addition, we found a positive correlation between the number of SGBs and the number of reflux episodes preceded by SGBs and the correlation between the number of SGBs and % AET at the distal esophagus. The patients with pH+ had significantly higher numbers of SGBs per day and the prevalence of reflux episodes preceded by SGBs compared to the pH– group. In contrast, a minority of reflux episodes were followed by belching. These findings elucidate the temporal relationship between SGBs and reflux episodes, and the role of SGB on distal esophageal acid exposure in GERD patients.

There have been limited studies exploring the association between reflux episodes and SGBs. In 2009, Hemmink et al⁶ performed a 24-hour esophageal pH-impedance monitoring in 50 patients with typical reflux symptoms and ten healthy controls. The prevalence of SGBs in patients with reflux symptoms was higher than that observed in healthy subjects, and 48.0% of SGBs were temporally related to reflux episodes (SGBs followed by reflux episodes 30.0% and SGBs preceded by reflux episodes 18.0%).⁶ In 2015, Koukias et al¹² conducted a retrospective study on 100 patients with SGBs more than 17 times/24-hour, showing that 41.0% of patients had positive pH monitoring. Notably, in a subgroup of patients with SGBs more than 124 times/24-hour, the prevalence of positive pH monitoring was 65.0%. They also found that patients with pH+ tended to have a higher number of SGBs, and 27.0% of the esophageal acid burden in these patients was due to reflux episodes preceded by SGBs.¹² Recently, in 2021, Jeong et al²⁰ conducted a retrospective study comparing belch characteristics in association with reflux between ten patients with typical GERD symptoms and 10 patients with belching disorders. They found that reflux-related SGB, including SGB during reflux and SGB that occurred immediately before the reflux onset, are more frequent in patients with GERD, while isolated belch is more frequent in patients with the belching disorder.²⁰ However, these studies could not clearly demonstrate the impact of SGBs on reflux episodes and the esophageal acid burden. Besides, gastroesophageal reflux characteristics in patients with SGB had never been extensively evaluated.

In our present study in GERD patients with significant SGBs, we compared reflux episodes and SGBs characteristics between patients with positive and negative esophageal pH monitoring. We found that about half of the reflux episodes had SGBs occurring just before reflux onset. The higher the numbers of SGBs, the higher the reflux episodes preceded by SGBs, and the higher the distal esophageal AET was demonstrated. Meanwhile, the patients with positive and negative esophageal pH monitoring had a similar proportion of the number of SGBs followed by reflux episodes to the total SGBs. Suppose we remove all reflux episodes preceded by SGBs from negative and positive pH test patients. In that case, the remaining numbers of reflux events (reflux episodes followed by SGBs and lone refluxes) in both groups will be similar, and all patients will have reflux parameters within the normal limit of the healthy volunteers.²¹ The fact that the remaining numbers of reflux episodes were normal and similar between patients with positive and negative pH monitoring suggests that significant distal esophageal acid exposure in GERD patients with SGB was only the result of SGBs aggravating more reflux events. (Fig. 2B, black parts of the bars). This study could not identify high BMI, low LES pressure, presence of hiatal hernia, or major esophageal motility disorders as risk factors for positive pH monitoring. This supports that SGBs induce significant esophageal acid exposure rather than classical GERD in nature in our patients in this study.

The mechanisms that explain how SGBs could elicit gastroesophageal reflux have been proposed as follows: (1) SGB-induced esophageal distension at the initial air-sucking phase may provoke transient lower esophageal sphincter relaxation (TLESR), (2) at the air-expelling phase, the intragastric pressure may overcome LES high-pressure zone during some episodes resulting in reflux, and (3) the TLESR-associated a reflex of upper esophageal sphincter relaxation may further induce air influx cycle.^13,22,23 All these mechanisms may relate to abnormal behavior. A prospective study by Glasinovic et al² in 31 patients with excessive SGBs showed that cognitive-behavioral therapy was associated with a significantly decreasing number of SGBs and acid reflux episodes, as well as the % AET at the distal esophagus.² A wait-list controlled cohort study by Ong et al⁵ in 36 patients with PPI-refractory GERD and SGB showed that a standardized diaphragmatic breathing exercise significantly reduced belching, GERD severity, and increased quality of life.⁵ These findings suggest the secondary GERD from excessive belching and treatment of SGB may be beneficial for decreasing distal esophageal acid exposure. Further randomized control study on behavioral therapy to reduce esophageal acid exposure is needed.

Previous studies suggested belching may occur in response to reflux or unpleasant feelings in the esophagus.^6,10-12 Although we found that reflux episodes followed by SGBs were significantly associated with acid reflux, only 1.8% (0.3-4.4%) of SGB episodes developed immediately after reflux episodes. So, acid suppressant in this group of patients is unlikely to improve SGB because most SGB episodes were not induced by reflux.

Another interesting point in our study is that reflux episodes preceded by SGBs and reflux episodes followed by SGBs extended more proximal than lone refluxes and had longer acid contact time and BCT. The proximal extent of reflux probably results from a balance between characteristics of the gastric contents, compliance of the esophagogastric junction, gastroesophageal pressure gradient, physical properties of the refluxate, and esophageal body tone or contraction.²⁴ The previous study could not demonstrate that gas in the refluxate influences the proximal extent of liquid reflux.²⁵ During the SGB, rapid air propagation in the esophagus may increase esophageal body tone.^24,26 Patients with a more tonic esophagus may have more proximal reflux migration and have longer BCT. In addition, SGB occurring immediately after reflux onset may cause re-reflux and prolonged BCT compared to lone reflux. The degree of the proximal extent of refluxate might determine esophageal and extraesophageal-reflux symptoms perception.^27-29 Patients with non-erosive reflux disease had more perception of proximal reflux.²⁷ In this study, we enrolled only patients without severe reflux esophagitis. Although our data suggest that SGB elimination in patients with SGB-associated GERD will return the patients to normal physiologic status or cure the GERD, further study on the effect of SGB treatment on esophageal and extraesophageal-reflux symptoms is needed.

The strength of this study is the relationship between SGBs and reflux episodes was clarified, and the characteristics of SGB were thoroughly explored by using 24-hour impedance-pH monitoring. Nevertheless, our study has several limitations. Firstly, this is a retrospective study. Other factors related to GERD, such as delayed gastric emptying and different meal calories, could not be evaluated. Secondly, all patients in this study had mild reflux esophagitis (Los Angeles grade A or B) or no esophagitis, so the results may not generally apply to all GERD phenotypes. A further well-designed prospective randomized study with larger samples of patients with pathologic reflux is needed to confirm the negative impact of SGB and the benefit of SGB management in GERD patients with SGB.

Conclusionally, in GERD patients with SGB, the higher the number of SGBs, the higher the reflux episodes preceded by SGBs is demonstrated. The higher number of SGBs per day is significantly associated with positive pH monitoring. Furthermore, the SGB associated reflux episodes may be related to prolonged reflux events. Identification and management of SGB may be beneficial in GERD patients with excessive belching by reducing the reflux episodes preceded by SGBs.

This article was presented in abstract form at The 6th Biennial Congress of the Asian Neurogastroenterology and Motility Association in April, 2019, Seoul, Korea.

This study was supported by the Ratchadapiseksompotch Fund, Chulalongkorn University. They had no roles in the study design or preparation of data, or the manuscript.

None.

Suriya Keeratichananont was involved in the study’s conception, design, analysis, and manuscript drafting; Tanisa Patcharatrakul was involved in the study’s conception, design, analysis, and critical revision of the manuscript; and Sutep Gonlachanvit was involved in the study’s conception, design, analysis, supervision, and critical revision of the manuscript. All authors read and approved the final manuscript.