Gastroesophageal reflux disease (GERD) is a common gastrointestinal disorder that causes significant psychological and socioeconomic burdens worldwide.¹ Proton pump inhibitors (PPIs) are the standard treatment choice for GERD. Many patients require long-term use of PPIs, and up to 40% do not respond completely to PPIs.²

Endoscopic management has emerged to fill the solution between medical and surgical treatments of GERD.³ Anti-reflux mucosal intervention (ARMI) has gained popularity in recent years due to increasing evidence in effectiveness for PPI-refractory or PPI-dependent GERD.^4-6

Among the ARMI methods, anti-reflux mucosal ablation (ARMA) ablates the mucosa, creates shrinking scarring to rebuild esophagogastric junction (EGJ) barrier, and the effect was proven up to 3 years in long-term observation.^6,7 However, potential negative effects of ARMA have also been observed and postulated to be associated with the change in motility from scarring.⁸ Moreover, previous studies mostly focused on the improvement of symptoms, esophageal acid exposure, and reflux episodes, while very few of them studied the motility changes post-ARMA.

Therefore, in the present study, with the help of high-resolution impedance manometry (HRIM) and the updated Chicago classification version 4.0, we first aimed to examine the detailed motility changes after ARMA. We furthermore tried to understand the influence of motility alterations on treatment outcomes.

Study Participants

Subjects with PPI-dependent GERD and intention to receive ARMA after shared decision-making with attending physicians between April 2021 and December 2022 were prospectively enrolled to join this study with comprehensive pre-ARMA and post-ARMA esophageal motility evaluation. All participants had long-term PPI use over 6 months and exhibited worsening symptoms after discontinuing PPI treatment for 2 weeks. Subjects were considered with GERD based on the following criteria: (1) acid exposure time (AET) exceeding 4% measured through pH monitoring off PPI therapy,^1,9 (2) numbers of reflux episodes (NRE) > 40 detected by impedance monitoring,¹ or (3) erosive esophagitis with Los Angeles classification grade A or higher documented by esophagogastroduodenoscopy. Our rationale of the inclusion criteria was based on the Montreal classification and Rome IV criteria (erosive esophagitis), and modified from Lyon Consensus 1.0 (pH and impedance criteria), while we also took into consideration AET cutoff > 4% as abnormal for Asian and Chinese populations.^9,10 Since the modern diagnosis of GERD has become more precise and rigorous during our study period, we have also re-evaluated our cases according to the latest Lyon Consensus 2.0 and Asian GERD consensus.^9-11 The study excluded individuals with significant comorbidities such as (1) liver cirrhosis, (2) malignancy, (3) congestive heart failure, (4) end-stage renal disease, (5) bleeding tendency, (6) pregnancy, (7) major esophageal motility disorders, such as achalasia, and (8) sliding hiatal hernias with a sac larger than 3 cm.⁷

The Research Ethic Committee of the National Taiwan University Hospital approved this study (Approval No. 202003029RINC). Informed consent was obtained from all participating patients after we thoroughly explained the advantages and potential drawbacks of the procedure.

Pre- and Post-procedural Evaluation

Upon acceptance of the study protocol, demographics and clinical information of the study participants were systematically collected, including age, gender, underlying medical conditions, weight and height, detailed GERD history, and current medications. Comprehensive standardized questionnaires, including GerdQ, reflux symptom index (RSI), and patient assessment of upper gastrointestinal symptom severity index (PAGI-SYM), as well as esophagogastroduodenoscopy, 24-hour multichannel intraluminal impedance-pH (MII-pH) monitoring, and HRIM results were collected and analyzed before and 3 months after ARMA.

Standardized Symptom Questionnaires

Comprehensive standardized questionnaires included GerdQ, RSI, and PAGI-SYM. GerdQ is a 6-item questionnaire for the diagnosis and management of GERD, consisting of 4 positive (heartburn, regurgitation, sleep disturbance due to reflux symptoms, and use of over-the-counter medication) and 2 negative (epigastric pain and nausea) predictors of GERD.¹² RSI is a 9-item outcomes instrument for atypical GERD symptoms, especially laryngeal symptoms.¹³ Twenty-item PAGI-SYM includes 6 subscales: heartburn/regurgitation, fullness/early satiety, nausea/vomiting, bloating, upper abdominal pain, and lower abdominal pain.¹⁴

Endoscopic Evaluation

The distal esophagus and EGJ were observed carefully during the endoscopic assessment. Erosive esophagitis was recorded according to the Los Angeles classification system.¹⁵ In brief, grade A esophagitis indicates 1 (or more) mucosal break(s) no longer than 5 mm that does not extend between the tops of 2 mucosal folds; grade B esophagitis indicates 1 (or more) mucosal break(s) longer than 5 mm, that does not extend between the tops of 2 mucosal folds; grade C esophagitis indicates 1 (or more) mucosal break(s) that are continuous between the tops of 2 or more mucosal folds but involve less than 75% of the esophageal circumference; grade D esophagitis indicates 1 (or more) mucosal break(s) involving at least 75% of the esophageal circumference. The gastroesophageal flap valve (GEFV) was evaluated using Hill’s classification.¹⁶

High-resolution Impedance Manometry

Prior to ARMA, the gastroenterologist (P.H.T.) conducted HRIM using a solid-state system (Laborie/MMS, Enschede, The Netherlands) to measure esophageal functionality and collected the data. The results were corroborated by an experienced pediatric gastroenterologist (J.F.W.). The final diagnosis was based on the updated Chicago classification version 4.0.¹⁷

Locations of the lower esophageal sphincter (LES) were determined by observing high-pressure zones in the esophagus^’ lower regions. The integrated relaxation pressure (IRP) and EGJ contractile integral (EGJ-CI) of the LES indicate its resting tone or pressure level. The distal contractile integral (DCI) is used to quantify the peristaltic activity in the distal esophagus.¹⁷ Provocative tests, including multiple rapid swallows (MRS) and rapid drinking challenge, were routinely performed and the results were recorded.

24-Hour Multichannel Intraluminal Impedance-pH Monitoring

Prior to ARMA, a 24-hour MII-pH monitoring evaluation (Laborie/MMS) was performed in each patient following a cessation of PPI usage for at least 2 weeks. The impedance-pH catheter was inserted transnasally, with a single electrode positioned 5 cm proximal to the LES. The analysis of all outcomes was performed by the same gastroenterologist (P.H.T.) via software with the impedance-pH recording apparatus (Laborie/MMS).

The parameters obtained from pH monitoring included acid exposure time (AET, the percentage of time that esophageal acid exposure was below a pH of 4.0), total number of reflux episodes, number of reflux episodes lasting over 5 minutes, duration of the longest reflux episode, and DeMeester score. Reflux events recorded from the impedance monitoring (NRE) were further divided into 3 categories based on the pH level at the time of reflux: acid reflux with pH < 4, weakly acidic reflux with pH 4-7, and weakly alkaline reflux with pH > 7. We also calculated a novel diagnostic parameter called the mean nocturnal baseline impedance (MNBI), defined as an average of baseline impedance at the 3-cm and 5-cm impedance channel at 3 time points during night sleep (around 1 AM, 2 AM, and 3 AM) for 10 minutes.¹⁸

Anti-reflux Mucosal Ablation

Patients were provided sedation through intravenous anesthesia by experienced anesthesiologists to ensure comfort during the ARMA procedures. A single skilled endoscopist (C.C.C.) conducted all the ARMA procedures using a gastroscope (GIF 260J; Olympus, Tokyo, Japan) with a transparent hood. The stomach was insufflated with absorbable CO₂, and a 320-degree horseshoe-shaped ablation was created on the lesser curvature side of the cardia, with an area of spared mucosa of 1-scope diameter on the greater curvature side of the cardia. The ablated mucosa on the lesser curvature side was wider (2-scope diameter) than the rest (1-scope diameter). The mucosal ablation was done with argon plasma coagulation (APC probe 20132-221, Vio 300D, Erbe, Tübingen, Germany; forced APC 100W, flow rate 1.0 L/min). Following the intervention, a careful post-procedural assessment was conducted to record any immediate adverse events, including pain, bleeding, or fever. The duration of hospitalization following the procedure was also carefully documented.

After the procedure, PPI was administered for 1 month to ensure ablation wound healing and scar formation, then PPI discontinuation was tried 1 month after the procedure if the patient could tolerate it. The usage of PPI and any associated clinical symptoms were recorded for the maximal duration. A comprehensive re-evaluation was conducted 3 months after ARMA with at least 2-week PPI discontinuation.

Statistical Methods

The statistical analysis entailed presenting categorical and continuous variables as percentages and mean with standard deviation and median with interquartile range for comparison of parameter changes after ARMA. The association between groups and categorical variables was assessed using a chi-squared test. A paired t test and McNemar’s test were used for comparing continuous data before and after ARMA. Mann–Whitney tests were employed for comparing continuous data between groups with different treatment responses. All statistical analyses were conducted using SPSS for Windows version 29.0 (IBM, Armonk, NY, USA).

Demographics

A total of 20 patients with PPI-dependent GERD who underwent ARMA were analyzed. The demographics and clinical characteristics are displayed in Table 1. The mean age was 50.6 ± 11.4 years and 80% were male. These patients presented with GerdQ and RSI scores of 11.1 ± 2.6 and 11.6 ± 5.5, respectively. The subjects were diagnosed with GERD based on the following criteria: 1 with erosive esophagitis LA grade B, 6 with AET > 6%, 1 with AET 4-6% and positive symptom association, 2 subjects with both AET 4-6% and NRE 40-80, 9 with NRE > 80, and 1 with NRE 40-80 and low MNBI. Approximately 60% of patients had AET ≥ 4%, and 65% had NRE ≥ 40. Eleven (55%) patients exhibited erosive esophagitis, with 10 patients (50%) categorized as LA grade A and 1 patient (5%) as LA grade B. With respect to Hill valve status, 1 patient (5%) presented with GEFV grade II, while the remaining patients (95%) presented with GEFV grade III. Upon HRIM analysis, 3 patients (15%) were diagnosed with ineffective esophageal motility (IEM), and 1 patient (5%) had absent contractility. The rest had relatively normal esophageal motility.

Table 1 . Demographics and Clinical Characteristics of Study Subjects

Characteristics	Total (n = 20)
Age (yr)	50.6 ± 11.4
20-40	4 (20)
40-60	10 (50)
> 60	6 (30)
Gender
Male	16 (80)
Female	4 (20)
Comorbidities
Hypertension	7 (35)
Diabetes	2 (10)
Dyslipidemia	5 (25)
Coronary artery disease	2 (10)
Smoking	2 (10)
Drinking	3 (15)
BMI (kg/m2)	23.8 ± 3.7
GERD symptom questionnaire
GerdQ score	11.1 ± 2.6
RSI score	11.6 ± 5.5
MII-pH
AET (%)	5.84 ± 4.63
> 6%	7 (35)
4-6%	5 (25)
< 4%	8 (40)
DeMeester score	19.85 ± 17.00
NRE	73.05 ± 19.34
Endoscopic findings
Erosive esophagitis	11 (55)
A	10 (50)
B	1 (5)
GEFV
II	1 (5)
III	19 (95)
HRIM diagnosis
Normal	16 (80)
IEM	3 (15)
Absent contractility	1 (5)

BMI, body mass index; GERD, gastroesophageal reflux disease; RSI, reflux symptom index; MII-pH, multichannel intraluminal impedance-pH; AET, acid exposure time; NRE, number of reflux episodes; GEFV, gastroesophageal flap valve; HRIM, high-resolution impedance manometry; IEM, ineffective esophageal motility.

Data are presented as mean ± SD or n (%).

Procedure-related Parameters and Adverse Events

As shown in Table 2, the mean procedure duration was 32.8 ± 15.1 minutes, and all procedures achieved complete ablation of the targeted mucosal areas. The average length of hospital admission after ARMA was 2.1 ± 0.7 days. Post-procedural fever was observed in 6 patients (30%). Four cases were resolved with antipyretics, while the other 2 were resolved with short-term intravenous antibiotics. One patient experienced tarry stool that subsided after conservative medical treatment. Dysphagia was reported by 6 patients (30%), and 2 of them (10%) underwent endoscopic balloon dilatation leading to symptom resolution. The others had gradual symptomatic improvement without any further intervention.

Table 2 . Anti-reflux Mucosal Ablation Procedure-related Parameters

Procedure-related parameter	Total (N = 20)
Mean procedure duration, minutes	32.8 ± 15.1
Hospital stay post-ARMA, days	2.1 ± 0.7
Post-ARMA complication
Fever	6 (30)
Dysphagia	6 (30)
Stenosis required balloon dilation	2 (10)
Bleeding	1 (5)

ARMA, anti-reflux mucosal ablation.

Data are presented as mean ± SD or n (%).

Effects of Anti-reflux Mucosal Ablation on Clinical Symptoms

The self-reported questionnaires shown in Table 3 revealed significantly lower overall GerdQ scores (11.16 ± 2.67 to 9.11 ± 2.64, P = 0.026) and RSI scores (11.63 ± 5.62 to 6.11 ± 3.86, P = 0.001). For GerdQ, the score of the item ‘sleep disturbance from GERD’ was significantly lower. The score of hoarseness, clearing throat, cough after eating or lying, and heartburn sensation on RSI were also significantly improved. In general, 12 (60%) of patients exhibited an improvement in their GerdQ scores, while 9 (45%) showed improvements in their RSI scores. Several PAGI-SYM symptom items improved significantly. Thirteen patients (65%) reported discontinuation of PPI during a follow-up up to 1 year.

Table 3 . Comparison of Symptom Profiles Before and 3 Months After Anti-reflux Mucosal Ablation

Questionnaire	Pre-ARMA	Post-ARMA	P-value
GerdQ
1. Heartburn	1.84 ± 1.46	1.00 ± 1.15	0.061
2. Regurgitation	2.21 ± 1.23	1.63 ± 1.26	0.157
3. Epigastralgia	2.32 ± 1.25	2.74 ± 0.81	0.163
4. Nausea	2.84 ± 0.69	2.84 ± 0.69	1.000
5. Sleep disturbance	1.95 ± 1.31	0.84 ± 1.26	0.015
6. OTC medications	0.00 ± 0.00	0.05 ± 0.23	0.331
Total scores	11.16 ± 2.67	9.11 ± 2.64	0.026
RSI
1. Hoarseness	2.16 ± 1.61	0.58 ± 0.96	< 0.001
2. Clearing throat	1.74 ± 1.63	0.95 ± 1.27	0.035
3. Excess throat mucus or postnasal drip	1.11 ± 1.41	0.79 ± 1.18	0.331
4. Difficulty swallowing food, liquid, or pills	0.63 ± 0.96	0.26 ± 0.65	0.202
5. Cough after eating or lying	0.63 ± 0.96	0.16 ± 0.50	0.046
6. Breathing difficulties or choking	0.74 ± 0.99	0.16 ± 0.50	0.053
7. Troublesome or annoying cough	0.00 ± 0.00	0.16 ± 0.50	0.187
8. Lump in throat	1.42 ± 1.46	0.89 ± 1.10	0.228
9. Heartburn, chest pain, indigestion, or stomach acid coming up	3.21 ± 1.27	2.16 ± 1.12	0.012
Total scores	11.63 ± 5.62	6.11 ± 3.86	0.001
PAGI-SYM
1. Heartburn during the day	1.84 ± 1.71	1.05 ± 1.35	0.144
2. Regurgitation during the day	1.89 ± 1.63	1.16 ± 1.34	0.172
3. Nausea	0.74 ± 1.24	0.58 ± 1.22	0.506
4. Upper abdominal pain	0.78 ± 1.35	0.50 ± 0.86	0.482
5. Stomach fullness	1.16 ± 1.46	0.47 ± 1.02	0.019
6. Loss of appetite	0.58 ± 1.12	0.37 ± 0.76	0.360
7. Upper abdominal discomfort	2.21 ± 1.40	1.16 ± 1.30	0.012
8. Bloating	2.11 ± 1.52	1.11 ± 1.33	0.006
9. Heartburn when lying down	2.05 ± 1.87	0.68 ± 1.42	0.019
10. Regurgitation when lying down	2.63 ± 1.89	1.63 ± 1.61	0.062
11. Lower abdominal pain	0.21 ± 0.63	0.05 ± 0.23	0.331
12. Feeling chest discomfort during the day	2.11 ± 1.59	0.89 ± 1.29	0.028
13. Bitter, acid, or sour taste in the mouth	2.21 ± 1.44	0.79 ± 1.13	0.002
14. Lower abdominal discomfort	0.21 ± 0.63	0.05 ± 0.23	0.331
15. Feeling chest discomfort at night	2.05 ± 1.72	0.47 ± 1.17	0.007
16. Retching	0.16 ± 0.69	0.21 ± 0.92	0.331
17. Belly visibly larger	1.16 ± 1.34	0.53 ± 0.96	0.048
18. Vomiting	0.42 ± 1.02	0.37 ± 1.12	0.841
19. Unable to finish a normal-sized meal	0.63 ± 1.16	0.53 ± 0.96	0.682
20. Feeling excessively full after meals	1.84 ± 1.57	0.68 ± 1.11	0.002

ARMA, anti-reflux mucosal ablation; OTC, over-the-counter; RSI, reflux severity index; PAGI-SYM, patient assessment of upper gastrointestinal symptom severity index.

Data are presented as mean ± SD.

Effects of Anti-reflux Mucosal Ablation on Endoscopic Findings

A remarkable 16 (80%) of subjects showed improvements in their GEFV status. Before ARMA, 19 patients (95%) had GEFV grade III. After ARMA treatment, GEFV grade I was noted in 6 patients (30%), grade II in 10 patients (50%), and grade III in only 4 patients (20%) (P < 0.001 with Yates^’ correction).

24-Hour Multichannel Intraluminal Impedance-pH Monitoring

The 24-hour MII-pH monitoring, as shown in Table 4, revealed a significantly lower AET (5.84 ± 4.63% vs 2.83 ± 3.41%, P = 0.024), especially in the upright posture (6.67 ± 5.71% vs 3.52 ± 3.96% P = 0.041). The number of acid refluxes, number of refluxes longer than 5 minutes, and DeMeester score also significantly decreased.

Table 4 . Comparison of 24-Hour Multichannel Intraluminal Impedance-pH Monitoring Before and 3 Months After Anti-reflux Mucosal Ablation

Parameter	Pre-ARMA	Post-ARMA	P-value
AET (%)	5.84 ± 4.63	2.83 ± 3.41	0.024
Upright	6.67 ± 5.71	3.52 ± 3.96	0.041
Supine	4.59 ± 7.68	1.6 ± 4.14	0.113
No. of reflux	56.95 ± 31.55	25.55 ± 18.35	0.001
No. of reflux > 5 min	3.65 ± 4.81	1.18 ± 2.01	0.046
Longest reflux (min)	11.77 ± 11.45	11.16 ± 18.47	0.859
DeMeester score	19.85 ± 17	9.73 ± 11.84	0.047
MNBI (5 cm), Ω	2069.79 ± 798.00	2359.05 ± 913.09	0.158
MNBI (3 cm), Ω	1775.74 ± 736.12	2055.05 ± 759.88	0.156
NRE	73.05 ± 19.34	37.55 ± 22.71	<0.001
Acidic reflux	46.6 ± 14.99	23.5 ± 15.86	<0.001
Liquid	9.26 ± 8.2	3.58 ± 3.66	0.011
Mixed	36.16 ± 14.71	20 ± 14.63	<0.001
Weakly acidic reflux	23.9 ± 11.96	12.85 ± 11.31	0.004
Liquid	4.95 ± 5.08	2.84 ± 4.46	0.151
Mixed	19.68 ± 10.77	10.05 ± 9.16	0.002
Weakly alkaline reflux	2.55 ± 6.27	1.2 ± 2.38	0.391
Liquid	0.63 ± 1.46	0.68 ± 1.86	0.921
Mixed	2.05 ± 5.09	0.58 ± 0.96	0.243
Gas reflux	38.7 ± 42.9	44.3 ± 35.06	0.500
Reflux extent (%)
17 cm	20.37 ± 11.47	13.95 ± 12.78	0.020
15 cm	34.42 ± 14.49	32.53 ± 16.23	0.636
9 cm	75.63 ± 9.86	78.21 ± 12.94	0.422
7 cm	86.79 ± 6.05	85.53 ± 12.61	0.673
5 cm	100	100	-
3 cm	100	100	-

ARMA, anti-reflux mucosal ablation; AET, acid expose time; MNBI, mean nocturnal baseline impedance; NRE, number of reflux episodes.

Data are presented as mean ± SD or n.

On impedance monitoring, the total number of reflux episodes decreased significantly after ARMA (73.05 ± 19.34 vs 37.55 ± 22.71, P < 0.001). A lower number of acid refluxes (46.6 ± 14.99 vs 23.5 ± 15.86, P < 0.001) and weak acid refluxes (23.9 ± 11.96 vs 12.85 ± 11.31, P = 0.004) were also noted. The proportion of the proximal reflux extent (17 cm above LES) also decreased post-ARMA (20.37 ± 11.47% vs 13.95 ± 12.78%, P = 0.020).

Esophageal Motility Changes After Anti-reflux Mucosal Ablation

As shown in Figure 1, 3 months after ARMA, significantly increased LES 4-second IRP (IRP-4s) (5.75 ± 6.42 mmHg to 9.99 ± 5.89 mmHg, P = 0.020), LES resting pressure (13.89 ± 10.78 mmHg to 21.68 ± 11.5 mmHg, P = 0.034), and EGJ-CI (16.42 ± 16.93 mmHg·cm to 31.95 ± 21.25 mmHg·cm, P = 0.016) were observed. Moreover, DCI also increased significantly from 966.85 ± 845.84 mmHg·s·cm to 1198.8 ± 811.74 mmHg·s·cm (P = 0.023).

Figure 1. Comparison of parameters of high-resolution impedance manometry before and 3 months after antireflux mucosal ablation (ARMA). Abbreviations: EGJ, esophagogastric junction; LES, lower esophageal sphincter; IRP-4s, 4-second integrated relaxation pressure.

After ARMA, 80% of patients had improvements in DCI, 65% had improvements in EGJ-CI, and 70% showed improvements in LES resting pressure. Two of the 3 patients with IEM had their abnormal esophageal motility reversed to normal, while 1 case of absent contractility improved to IEM (Fig. 2). However, 1 patient with normal esophageal motility and a sliding hiatal hernia developed IEM after the procedure.

Figure 2. Endoscopic findings and high-resolution impedance manometry (HRIM) in a representative case before and 3 months after antireflux mucosal ablation (ARMA). (A) Wide cardia opening (gastroesophageal flap valve grade III) before ARMA. (B) Tightened cardia (gastroesophageal flap valve grade I) after ARMA. (C) Absent contractility (100% failed contraction) on HRIM before ARMA. (D) Improved esophageal motility (10% intact, 40% weak, and 50% failed contraction) after ARMA.

Before ARMA, 6 out of 20 patients showed negative MRS, but none of them developed dysphagia after ARMA. Instead, 5 of them had esophageal motility improvement with MRS changing from negative to positive results and experienced improved symptoms on the GerdQ after ARMA. For the 6 patients who had dysphagia after ARMA, none of them had negative MRS before ARMA. None of our study subjects had abnormal findings on rapid drinking challenge before or after ARMA.

Factors Predicting Symptom Improvements After Anti-reflux Mucosal Ablation

We further compared the physiological parameters (both Pre-ARMA and post-ARMA) between subjects with and without symptom improvement, as defined by improvement of GerdQ items 1 plus 2, after ARMA (Table 5). Regarding the parameter changes after ARMA, NRE was decreased to a higher extent in the symptom improvement group (–47.00 [–53.50, –27.50] vs –33.00 [–44.00, –19.00], P = 0.044). MNBI 3 cm (659 [200.5, 937] vs –405.5 [–543.5, –225.25], P = 0.017) and MNBI 5 cm (731 [418, 1017.5] vs –486.5 [–758.75, 137.25], P = 0.008) increased more significantly in the symptom improvement group. Patients with symptom improvement had better pre-AMRA body motility in terms of the proportion of intact peristalsis (100 [90, 100]% vs 50 [40, 80]%, P = 0.042) and mean DCI (1244 [799.5, 1694] mmHg·s·cm vs 523 [302, 600] mmHg·s·cm, P = 0.014).

Table 5 . Comparison of 24-Hour Multichannel Intraluminal Impedance-pH and High-resolution Impedance Manometry Parameter Between Patients Who Had Symptom Improvement on Gastroesophageal Reflux Disease Questionnaire

Parameter	Improvement (n = 11)	No improvement (n = 9)	P-value
Post-ARMA parameter change
AETa	–1.80 (–3.7, –1.00)	–2.1 (–3.3, 1.2)	0.909
NREa	–47 (–53.5, –27.5)	–33 (–44, –19)	0.044
MNBI 3cma	659 (200.5, 937)	–405.5 (–543.5, 225.25)	0.017
MNBI 5cma	731 (418, 1017.5)	–486.5 (–758.75, 137.25)	0.008
DCIa	152 (40.5, 694.5)	296 (168, 431)	0.970
EGJ-CIa	8 (–4.5, 18)	20 (9.25, 34.75)	0.282
LES Resting Pressurea	4 (–1.5, 11.5)	12 (4, 21.25)	0.265
Pre-ARMA parameters
AET	3.8 (2.40, 7.1)	4.4 (4.2, 9.3)	0.403
NRE	81 (55.00, 85.5)	79 (53, 86)	0.97
MNBI 3cm	1690 (1252, 2318)	1893 (1026, 2539)	0.953
MNBI 5cm	1730 (1460, 2660)	2349 (1225, 2820)	0.937
Intact peristalsis	100 (90, 100)	50 (40, 80)	0.042
Weak peristalsis	0 (0, 5)	30 (0, 40)	0.03
Failed peristalsis	0 (0, 0)	10 (0, 30)	0.078
DCI	1244 (799.5, 1694)	523 (302, 600)	0.014
EGJ-CI	12 (9, 18)	8 (2, 32)	0.917
LES Resting Pressure	12 (9, 22)	6 (4, 13)	0.183

^aCalculated by subtracting the difference of before and after the anti-reflux mucosal ablation (ARMA) procedure.

AET, acid exposure time; NRE, number of reflux episodes; MNBI, mean nocturnal baseline impedance; DCI, distal contractile integral; EGJ-CI, esophagogastric junction contractile integral; LES, lower esophageal sphincter.

Data are presented as median (interquartile range).

Our study demonstrates that ARMA improved GERD symptoms, leading to a high rate of discontinuing PPI use. According to the SRQ assessment, a notable 60% of patients demonstrated improvement in their GerdQ scores, and an impressive 45% exhibited positive changes in their RSI scores. Moreover, 13 patients (65%) were able to discontinue their PPI use, marking a significant shift in their treatment regimen. For those with erosive esophagitis, 30% experienced amelioration, while a remarkable 80% showed improvement in their GEFV status.

In the present study, 6 patients developed dysphagia after ARMA. Previous studies have suggested negative MRS may help to predict the development of post-fundoplication dysphagia in GERD patients with IEM.¹⁹ Nevertheless, in the present study, none of the subjects with a negative MRS developed dysphagia after ARMA. For patients who had dysphagia after ARMA, none of them had negative MRS before ARMA. Our findings suggest a poor correlation between the MRS (esophageal reserve) and the development of dysphagia after ARMA. However, since our case number was relatively small, further large-scale studies on the role of MRS in the ARMA or other ARMI procedures are warranted.

It is worth highlighting that our cases revealed esophageal motility improvement, including both EGJ and body contractility. Up to 80% of patients had improvements in DCI, 65% had improvements in EGJ-CI, and 70% had improvements in LES resting pressure. This result may help to clarify the role of esophageal motility in the mechanism of GERD.

The mechanism behind GERD is complex, and long-term treatment with PPIs often fails to address its underlying pathophysiology.^2,20,21 Comprehensive analysis based on the latest HRIM and MII-pH monitoring before and after ARMI can help us understand GERD and anti-reflux mechanisms. Based on objective MII-pH monitoring, we demonstrated that 65% of patients had improved AET and 85% had improved reflux episodes. Moreover, 65% of all patients exhibited an increase in MNBI, suggesting improvement of esophageal mucosal barrier function. Remarkably, 2 patients with IEM were reversed to normal and 1 patient with absent contractility improved to IEM. This demonstrated that ARMA can improve GERD through not only barrier reconstruction but also esophageal motility improvement.

Another important finding of our study was the improvement in various atypical symptoms, which may be attributed to the substantially improved esophageal motility in GERD with better acid clearance and shortened reflux period. Decreased severity of hoarseness, throat clearing, and cough after eating or lying represented less laryngopharyngeal reflux with or without reflex symptoms triggered by the vagal reaction.²² Our study consistent with previous research in observed improvement of DCI, indicating the recovery of the contractile function of the esophageal body helps in acid clearance and bolus transit.^23,24 Ribolsi et al²⁵ indicated that impaired esophageal peristalsis may account for delayed bolus transit and reduced esophageal reflux clearance in patients with GERD, and large peristaltic breaks are often identified in GERD-related chronic cough. However, research on detailed esophageal motility evaluation after anti-reflux treatment is limited. Our study demonstrates remarkable enhancement of EGJ barrier function post-procedure, as indicated by LES IRP-4s, LES resting pressure and EGJ-CI.^24,26

Riva et al²⁷ showed 36% of patients with IEM reversed to normal esophageal motility after the Linx procedure. Our investigations also showed that 75% of patients with IEM or absent contractility had esophageal motility improvement after the ARMA procedure, consistent with previous results. Nevertheless, 1 patient experienced worsened motility after the procedure, together with increased esophageal acid exposure, EGJ-CI, LES resting pressure, and decreased DCI. Future studies to optimize the patient selection for ARMA procedures are warranted.⁴

ARMA has a positive effect on the LES barrier as indicated by LES IRP-4s, LES resting pressure, and EGJ-CI. The result may be partially explained by GEFV reconstruction and tightening of EGJ from the ARMA procedure. However, the actual mechanism underlying the esophageal contractility improvement after ARMA remains unclear, and we hypothesize that reduction of reflux burden and resolution of esophageal inflammation may contribute to the improvement of body motility function.²³ Earlier studies have shown that chronic esophageal acid exposure has been associated with ineffective esophageal motility.^28,29 In addition, Chen et al³⁰ noted that differences in esophageal bolus transit between healthy, NERD and erosive esophagitis groups might reflect a continuum of esophageal dysfunction secondary to increasing esophageal mucosal damage.³¹ Esophageal body motility improvement was also observed in other anti-reflux procedures, such as the Linx procedure.²⁷ Taken together, the reduction of the acid burden and associated esophageal inflammation by ARMA procedure may explain the improvement of the contractility of the esophageal body.

The study also highlights the ARMA instrument choice. The ARMA procedure can be performed using either a hybrid APC, APC or TT knife.⁴ Recent research by Shimamura et al³² (42.6% spray coagulation, 57.4% conventional APC) demonstrated a 60% clinical success rate, with 8.8% delayed bleeding, and 1.5% fever. Our study, using conventional APC, yielded satisfactory results with comparable adverse event rates, except for a relatively higher transient fever occurrence (30%), possibly linked to APC's thermal injury depth.^6,8

However, our study was subject to several limitations. First, this was a single-arm study without a randomized control group. The subjective outcomes and questionnaires could have been affected by placebo effect. Second, our relatively small sample size may obstruct the generalizability of the results. Third, the study only presents short-term outcomes. A larger-scale, long-term follow-up is needed to confirm the chronicity of the effect of ARMA on GERD. Fourth, regarding the study population, the modern diagnosis of GERD kept evolving during our study and analysis period. Therefore, we have re-evaluated our cases using the updated Lyon 2.0 and Seoul consensus.^9,11 Eighteen subjects (90%) fit the Lyon 2.0 criteria of proven GERD or borderline GERD plus at least 1 adjunctive evidence. The remaining 2 subjects with AET 4-6% and NRE 40-80 could still be considered as GERD based on the Asian GERD consensus.⁹ Therefore, we believe our study results could still be generalizable to the GERD population based on the latest GERD consensus. A comprehensive set of enrolment criteria, such as the actionable GERD proposed by Lyon 2.0, are required in future studies. Finally, since the ambulatory pH-impedance tests of our study were performed off PPI to evaluate the reflux burden, we could not exclude the possibility that some of the enrolled subjects with proven GERD may have overlap reflux hypersensitivity or functional heartburn as proposed by Rome IV.

In conclusion, ARMA effectively improved GERD, as indicated by PPI discontinuation, symptom relief, reduction in acid exposure, and impedance reflux improvement, during a short-term follow-up. Additionally, we demonstrated the concurrent evolution in EGJ barrier function and esophageal body contractility. Together, these results contribute to our understanding of the effects and mechanisms of ARMA on GERD. Longer follow-up to evaluate the sustainability of the effects of ARMA is warranted.

The authors would also like to thank Unit-Edit (www.uni-edit.net) for editing and proofreading this manuscript. The authors would also like to thank the staff of the Eighth Core Lab in the Department of Medical Research at the National Taiwan University Hospital for their technical support during this study.

This study was supported by research grants from the National Taiwan University Hospital (NTUH 112-S0320) and the Ministry of Science and Technology (MOST 111-2314-B-002-189-MY3). The funding agencies had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

None.

Chu-Kuang Chou, Chien-Chuan Chen, and Ping-Huei Tseng: conceptualization; Chu-Kuang Chou, Chien-Chuan Chen, Ming-Ching Yuan, Kun-Feng Tsai, and Jia-Feng Wu: data Curation; Chu-Kuang Chou, Chien-Chuan Chen, Ming-Ching Yuan, Kun-Feng Tsai, and Wei-Chi Liao: formal analysis; Chu-Kuang Chou, and Chien-Chuan Chen: writing of original draft; Han-Mo Chiu, Hsiu-Po Wang, Ming-Shiang Wu, and Ping-Huei Tseng: writing of review and editing; and Ping-Huei Tseng: supervision. All authors substantially contributed to the concept and design of the study, revised the article for important intellectual content, approved the final version for publication, and accountable for all aspects of the work.