Esophageal Reflux Hypersensitivity: A Comprehensive Review

Abstract

Reflux hypersensitivity (RH) is one of the phenotypes of gastroesophageal reflux disease. The latest Rome IV defines RH as a condition with typical reflux symptoms and positive reflux-symptom association despite normal acid exposure. Subsequently, the Lyon consensus proposed detailed cutoff values for the criteria on the basis of experts’ consensus. Rome IV brought a clear-cut perspective into the pathophysiology of gastroesophageal reflux disease and the importance of esophageal hypersensitivity. This perspective can be supported by the fact that other functional gastrointestinal disorders such as irritable bowel syndrome and functional dyspepsia often overlap with RH. Although several possible pathophysiological mechanisms of esophageal hypersensitivity have been identified, there is still unmet medical needs in terms of treatment for this condition. This review summarizes the current knowledge regarding RH.

INTRODUCTION

Reflux hypersensitivity (RH) is one of the functional esophageal disorders in Rome IV.¹ RH is defined as patients with typical reflux symptoms having normal acid exposure and positive reflux symptom association. The definition of RH, formerly called hypersensitive esophagus, changed with the revision of the Rome criteria. In the previous criteria (Rome III), RH and patients with abnormal acid exposure were encompassed within the same entity (i.e., nonerosive reflux disease [NERD]) in spite of inconsistency in response to acid suppression therapy and mechanisms of reflux symptom generation.² Subsequently, Rome IV introduced a new classification of gastroesophageal reflux disease (GERD) including a specific category called RH on the basis of the balance between two main pathophysiological parameters: esophageal acid exposure and esophageal hypersensitivity. For acid exposure, highly effective therapeutic options are available such as acid suppressive therapy and anti-reflux surgery (ARS). The healing of erosive esophagitis (EE) can be achieved in more than 90% of the cases by proton pump inhibitors (PPIs) or potassium-competitive acid blockers (PCABs).^3-5 However, these treatments cannot alleviate reflux symptoms enough in up to 48% of reflux patients.^6,7 The difference of treatment outcome between mucosal healing and reflux symptoms is probably attributable to esophageal hypersensitivity since there is poor correlation between the severity of mucosal damage and the severity of symptoms.^8-10 The process of symptom generation starts from the esophageal sensory nerve fibers stimulated by gastroesophageal reflux, followed by the transmission of the simulation to the central nervous system via vagal afferent nerves or spinal nerves. Although many studies found that there are multitude of central and peripheral factors related to esophageal hypersensitivity, this extremely complex process has not been completely elucidated, and makes the treatment for this condition challenging.

In this comprehensive review, we summarized the current knowledge about RH from epidemiology to treatment.

EPIDEMIOLOGY

RH affects a relatively small proportion of patients with GERD. However, the number of RH patients is not negligible considering the high global prevalence of GERD (14.8%).¹¹ Martinez et al.¹² studied patients with heartburn who were not yet treated with PPIs. They found that 9% of the reflux patients had RH, which accounts for 20% of patients with normal upper endoscopy (Fig. 1). This study could not identify nonacid reflux because of the lack of impedance sensors, then Savarino et al.⁸ investigated patients with typical reflux symptoms using off-PPI impedance-pH monitoring. The study showed similar proportion of RH, NERD, and functional heartburn (FH) in endoscopy-negative patients to the previous study when calculating reflux-symptom association only for acid reflux episodes. However, incorporation of nonacid reflux episodes into reflux-symptom association increased the prevalence of RH by approximately 10%. A similar diagnostic yield was seen in other off-PPI impedance-pH monitoring studies (Fig. 1).^13,14

Fig. 1.

Prevalence of each reflux phenotype in all patients with heartburn (A) and in endoscopy-negative patients with heartburn and/or regurgitation (B). The patients in both studies were phenotyped after at least a 14-day discontinuation of acid suppressive therapy. Only pH monitoring was used in (A), while impedance-pH monitoring was used in (B) so that the reflux-symptom association could be calculated for both acid and nonacid reflux episodes.

NERD, nonerosive reflux disease; RH, reflux hypersensitivity; FH, functional heartburn.

On-PPI impedance-pH monitoring showed that the RH prevalence in endoscopy-negative patients was similar to that observed in studies performed off-PPI. Roman et al.¹⁵ reported 36% of PPI nonresponder patients was diagnosed as RH on PPI. Patel et al.¹⁶ also found similar prevalence of RH between on- and off-PPI studies. However, part of NERD patients diagnosed off PPI can be converted into RH when tested on-PPI.¹⁷ Thus, RH cohort defined by off-PPI study is not necessarily same as those by on-PPI study.

The regional difference influences the RH prevalence. Several large Japanese on-PPI studies report that the prevalence of RH is up to 44%, which is higher than in Western countries.^18-20

DIAGNOSIS OF RH

The two experts’ consensuses, Rome IV criteria and the Lyon consensus, are the bedrock of RH diagnosis. Their proposed criteria for RH require dedicated esophageal physiology testing (i.e., manometry and impedance-pH monitoring). Thus, RH is normally diagnosed at tertiary centers due to the availability of such techniques. The Lyon consensus complements Rome IV criteria providing detailed cutoff values, specifically for impedance-pH monitoring (Fig. 2).

Fig. 2.

Diagnostic algorithm of PPI/PCAB refractory reflux patients.

PPI, proton pump inhibitor; PCAB, potassium-competitive acid blocker; SGB, supragastric belching; RS, rumination syndrome; EGD, esophagogastricduodenoscopy; HRM, high-resolution manometry; HRIM, high-resolution impedance manometry; RS, rumination syndrome; EoE, eosinophilic esophagitis; GERD, gastroesophageal reflux disease; LA, Los Angeles classification; LSBE, long segment Barrett’s esophagus; AET, acid exposure time.

1. Rome IV criteria

The latest Rome IV criteria classifies GERD into four phenotypes as follows: (1) EE, (2) NERD, (3) RH, (4) FH.¹ It is suggested that acid exposure predominantly accounts for esophageal symptoms in EE and NERD whilst esophageal hypersensitivity overrides acid reflux in RH and FH.

The diagnostic criteria of RH are as follows: (1) retrosternal symptoms including heartburn and chest pain; (2) normal endoscopy and absence of evidence that eosinophilic esophagitis is the cause for symptoms; (3) absence of major esophageal motor disorders (achalasia/esophagogastric junction outflow obstruction, diffuse esophageal spasm, jackhammer esophagus, absent peristalsis); and (4) evidence of triggering of symptoms by reflux events despite normal acid exposure on pH only or impedance-pH monitoring.

The last condition has been amended from Rome III as response to antisecretory therapy is no longer required for the diagnosis. All of the four items must be fulfilled for the past 3 months with symptom onset at least 6 months before diagnosis with a frequency of at least twice a week. In clinical practice, regurgitation is often seen as one of the typical symptoms in RH in spite of the criteria.

The diagnostic work-up incudes multiple biopsies from different levels of the esophagus to exclude eosinophilic esophagitis during endoscopy, high-resolution manometry and pH only or impedance-pH monitoring. High-resolution manometry is essential to exclude motility disorders and locate the upper edge of the lower esophageal sphincter to position esophageal pH sensor precisely.²¹ Typical reflux symptoms such as heartburn and/or regurgitation can be presented by up to 75%, 35% and 20% of patients with achalasia, hypercontractile esophagus and distal esophageal spasm respectively.^22-25

Rome IV also introduces the concept that RH and FH can overlap with proven GERD (i.e., Los Angeles classification grade B/C/D or long segment Barrett’s esophagus) as some patients still suffer from reflux symptoms after the mucosal healing.^1,26 This concept seems to be supported by the similar psychological profiles and the prevalence of other functional gastrointestinal disorders such as functional dyspepsia (FD) and irritable bowel syndrome (IBS) between RH/FH with and without proven GERD.²⁷ This kind of overlap between functional and organic disorders is also seen in bowel disorders as some patients with inflammatory bowel diseases suffer from IBS-like symptoms even in remission.²⁸

2. Lyon consensus

Although the Rome IV criteria defines RH, there was lack of concrete cutoff values for two conditions, (1) normal esophageal acid exposure and (2) positive symptom reflux association. The Lyon consensus addressed this issue where normal and pathological acid exposure time (AET) were proposed as <4% and >6%, respectively.²⁹ Concerning reflux-symptom association, symptom index (SI) and symptom association probability (SAP) are the two most common indices. In many cases, a symptom is considered related to reflux when the symptom is marked within 2 minutes after a reflux episode. SI is the proportion of reflux-related symptoms to the total number of reflux symptoms which considered positive when ≥50%. SAP uses a Fisher exact test for the presence or absence of reflux and/or symptoms for each 2 minutes segment of the entire tracings. A p<0.05 indicates the probability of chance association between reflux symptoms and reflux episodes is <5%, corresponding to a positive SAP (>95%).

The diagnosis of RH can be inconclusive with AET between 4% and 6% and/or only SI or SAP positive. There would be two options to make a robust diagnosis for RH as follows: (1) performing prolonged wireless pH monitoring, or (2) using other MII-pH parameters as a complement. The choice largely depends on local institutional expertise and availability of the techniques.

Wireless pH monitoring can record up to 96 hours, which decreases the influence of day-to-day variability of AET.³⁰ As even healthy subjects have pathological AET at times, RH patients can be diagnosed as NERD if tested with 24-hour pH metry on the worst reflux day. Hasak et al.³¹ found that longer recording time (e.g., 72 or 96 hours) can provide diagnostic yield especially when AET in the first 2 days are discordant or borderline. The complementary metrics include the number of reflux episodes, mean nocturnal baseline impedance (MNBI) and post-reflux swallow peristaltic wave (PSPW) index. The Lyon consensus proposes that >80 reflux episodes per 24 hours is conceivably abnormal whereas <40 is physiological. Recently <40 and <35 reflux episodes have been validated as physiological for heartburn and regurgitation respectively.^32,33 MNBI reflects cumulative damage of the mucosa, which is known as a surrogate marker for mucosal integrity (MI). Thus, little influence of day-to-day variability of acid reflux. PSPW index is the proportion of reflux episodes followed by saliva swallowing within 30 seconds from the end of reflux episode, which reflects the ability of esophageal chemical clearance.³⁴ In fact, several studies showed lower PSPW index in acid reflux predominant GERD phenotypes (i.e., EE and NERD) compared to esophageal hypersensitivity predominant.³⁴ Frazzoni et al.³³ showed 50% for PSPW index and 2,000 Ω for MNBI at 3 cm above the lower esophageal sphincter as optimal cutoff values to predict PPI-dependent heartburn compared to asymptomatic subjects. On the other hand, the largest healthy subjects’ data (n=391) of 24-hour off-PPI impedance-pH monitoring calculated a cutoff value of MNBI and PSPW index as 1,384 Ω and 15%, respectively.³⁵ These studies suggest that patients with >50% PSPW index and/or >2,000 Ω MNBI probably have normal acid exposure on average even when AET is inconclusive in 24-hour impedance-pH monitoring.

PATHOPHYSIOLOGY OF ESOPHAGEAL HYPERSENSITIVITY

Several studies demonstrated esophageal hypersensitivity to noxious stimuli in GERD patients. Trimble et al.³⁶ showed that patients with RH had a lower sensory threshold for esophageal balloon distension compared to those with NERD, Barrett’s esophagus or asymptomatic healthy subjects. Although the entire mechanism of esophageal hypersensitivity remains to be elucidated, several potential factors have been reported so far (Fig. 3).

Fig. 3.

Mechanism underlying esophageal reflux hypersensitivity. Exogenous stimuli perceived by the esophageal sensory nerves are transmitted to the central nervous system via two ways; vagal afferent nerves via the nodose ganglion and spinal nerves via dorsal root ganglion. Several factors contribute to reflux hypersensitivity influencing peripheral sensors in the esophagus or the central nervous system. Some mechanisms such as mucosal integrity, cytokines and prostaglandin E2, sensory receptors (TRPV1 and ASIC3) and superficial mucosal afferent nerves are associated with peripheral hypersensitivity. Other mechanisms such as esophageal hypervigilance along with anxiety and sleep disturbance mainly affect the processing of the conveyed stimuli in the central nervous system, which results in amplification of the sensation. These factors also might enhance peripheral sensitivity by impairing the mucosal integrity.

TRPV1, transient receptor potential vanilloid receptor-1; ASIC3, acid-sensing ion channel 3.

1. Mucosal integrity

Reflux symptoms can occur when the refluxate comes in contact with the esophageal mucosa. In EE, such stimuli including acid, bile, and pepsin can reach sensory afferent nerve endings in the esophageal wall via the mucosal deficit.³⁷ On the other hand, microscopic mucosal damage impairs MI in NERD. MI reflects barrier function of the mucosa, which can be measured in vivo (e.g. impedance monitoring) or in vitro (e.g. transepithelial electrical resistance, molecule permeability using ushing chamber, and transmission electron microscopy).³⁸ Impaired MI lets noxious substances pass through the mucosa, and MI inversely correlates to acid sensitivity in patients with NERD (i.e., lower MI more hypersensitive).^10,39 However, Weijenborg et al.⁴⁰ found no difference in MI between NERD and FH even though NERD had the shorter perception time in the acid perfusion test. These findings indicate that MI is not a sole factor to explain acid sensitivity.

Esophageal mucosal dilated intercellular space (DIS) is a histological condition associated with impaired MI. There is a causal link between DIS and MI as acid and bile perfusion cause both conditions simultaneously.^38,41 DIS can be found not only in EE and NERD but also in patients with normal AET.^37,42,43 Thus, DIS could explain the mechanism of symptom perception in RH. However, it should be noted that DIS can be found in up to 30% of asymptomatic subjects.⁴⁴ On top of that, the DIS is predominantly located in the basal layer, which questions whether it truly increases epithelial permeability for noxious substances from the luminal side.

2. Cytokines and prostaglandin E2

Despite the orthodox hypothesis that reflux damages esophageal mucosa directly from the luminal side, Souza et al.⁴⁵ suggests the opposite way of the development of esophageal inflammation by reflux (i.e., from inside the mucosa via cytokines). Their animal study found that acid reflux causes the epithelium to secrete proinflammatory cytokines such as interleukin (IL)-8 and IL1-b, and subsequently mainly IL-8 induces the migration of T cells and neutrophils to the submucosal layer, finally extending to the epithelial surface. Isomoto et al.⁴⁶ showed increased IL-8 expression in the esophageal epithelium of NERD patients, which lowered after 8-week PPI treatment in line with reflux symptom resolution. This implies the involvement of IL-8-induced mucosal inflammation in symptom generation.

Prostaglandin E2 (PGE2) also plays an important role in visceral hypersensitivity both at central and peripheral levels via the prostaglandin E2 receptor-1 (EP-1). Kondo et al.⁴⁷ showed that the level of PGE2 in the esophageal mucosa increased after acid perfusion which correlated to intensity of heartburn. EP1 receptor expressed in the mucosa and co-localized on protein gene product 9.5 immunoreactive nerve specifically in the muscularis mucosae and submucosa. In line with the results, EP-1 receptor antagonists and diclofenac (nonsteroidal anti-inflammatory drug) attenuated acid infusion induced heartburn or visceral hypersensitivity.^48-50

3. Superficial mucosal afferent nerves

Esophageal afferent sensory nerves are distributed in the esophageal mucosa and are likely to play an important role on perceiving noxious stimuli from the luminal surface. The afferent nerves are normally located closer to the lumen in the proximal esophagus compared to the distal esophagus.⁵¹ It might explain acid hypersensitivity in the proximal esophagus.⁵² Additionally, proximal reflux is one of the determinants triggering reflux symptoms.^53,54 Given these findings, the sensitivity of mucosal afferent nerves presumably depends on its proximity to the lumen. Our previous study found that NERD patients show more superficial mucosal afferent nerve (i.e., closer to the lumen) compared to healthy subjects.⁵⁵ Thus, the superficial nerve may partly underlie the mechanism of esophageal hypersensitivity. Interestingly, FH patients had mucosal afferent nerve at the similar depth to healthy subjects.⁵⁶ Esophageal hypersensitivity in FH might be predominantly formed by central factors rather than peripheral factors. Further study is warranted to investigate esophageal mucosal innervation of RH patients.

4. Sensory receptors (TRPV1 and ASIC3)

Transient receptor potential vanilloid receptor-1 (TRPV1) is a polymodal, nociceptive cation channel. TRPV1 can be activated by several exogenous factors including protons (H⁺), capsaicin and high temperature.⁵⁷ The expression of TRPV1 can be seen in not only afferent nerve fibers but also non-neural cells.^58-60 Guarino et al.⁶¹ found NERD and EE patients have higher mRNA and protein expression levels of TRPV1 in esophageal mucosal biopsies than healthy control. Besides, our recent study showed that TRPV1 is frequently expressed on superficial afferent nerves in the esophageal epithelium whereas occasional expression was seen on deep nerves around basal layer.⁶² On the other hand, a randomized controlled study by Krarup et al.⁶³ showed no inhibitory effect of TRPV1 antagonist on chest pain induced by heat, distension or electronic current in the esophagus. With regard to other TRPs receptors, transient receptor potential melastatin 8 (TRPM8) might be involved in esophageal hypersensitivity as menthol (i.e., TRPM8 agonists) infusion into the esophagus triggers different sensation in healthy subjects (cold sensation) and GERD patients (heartburn).⁶⁴

A few studies have reported the role of acid-sensing ion channels (ASICs) in human GERD. ASICs are voltage-insensitive epithelial Na+ channels responding to extracellular acidification, which function as nociceptor as well as mechanoreceptor.⁶⁵ Immunohistochemistry shows increased expression of ASIC3 in NERD compared to FH.⁶² On the other hand, mRNA expression level of ASIC3 in the esophageal mucosa does not differ between NERD and control.^62,66

Proteinase-activated receptor-2 (PAR-2) is a tethered ligand G protein-coupled receptor with seven transmembrane domains. Serine proteases including mast cell tryptase and pancreatic trypsin can activate PAR-2, which leads to visceral hypersensitivity and pain as a result of inflammatory and neuroinflammatory epithelial response via the secretion of IL-8, substance P and calcitonin gene-related peptide.^67,68 Kandulski et al.⁶⁷ found that PAR-2 and IL-8 expressed in the esophageal epithelium markedly in NERD and EE compared to control. Furthermore, these expressions correlated to histological changes such as DIS, basal cell hyperplasia and papillary elongation.

Wu et al.⁶⁹ studied the link between TRPV1, ASIC3 and PAR-2 using human esophageal epithelial cells and found that activation of PAR-2 enhances a pain-related substance release (i.e., acid-induced adenosine triphosphate) from the esophageal epithelium through TRPV1 and ASIC3.

5. Sleep disturbance

Sleep disturbance is one of the common coexisting conditions in GERD. In a U.S. nationwide survey of 1,000 patients with at least once-a-week heartburn, 79% of the patients reported nighttime heartburn, and 75% of whom considered the reflux symptoms affected their sleep.⁷⁰ As for a Japanese nationwide survey, 56.3% of 2,426 patients with heartburn had sleep disorders which was significantly higher than those without (40.7%), and there was a positive correlation between frequency of heartburn and the prevalence of sleep disturbance.⁷¹

GERD and sleep disturbance are likely to affect each other bidirectionally. Nighttime reflux provokes heartburn, which leads to difficulty in falling asleep, nocturnal and/or early morning awakening.⁷² Contrarily, sleep deprivation sensitizes the esophagus to acid and capsaicin (TRPV1 agonist).^73,74 It is of interest that this effect was seen in GERD patients, however not in healthy volunteers.⁷³ Sleep disturbance probably affects not only the processing of sensation in the central nervous system but also peripheral sensitivity since acute stress impaired mucosa integrity with DIS in animal study.⁷⁵ GERD and sleep disturbance can be exacerbated each other in a “vicious cycle” where GERD deteriorates sleep quality which in turn sensitizes the esophagus.⁷³ Our previous study found endoscopy-negative GERD was a significant risk factor of sleep disturbance compared to EE (odds ratio, 2.18; 95% confidence interval, 1.05 to 4.53),⁷⁶ which might underscore the influence of sleep disturbance towards RH rather than the opposite direction.

6. Esophageal hypervigilance and anxiety

It is well known that psychological factors such as anxiety and depression contribute to symptom generation as well as its intensity in many conditions. This is the case in reflux symptoms as there is a higher level of visceral hypersensitivity, anxiety and depression across endoscopy-negative reflux phenotypes (i.e., RH, FH, and NERD) compared to healthy control.⁷⁷ Moreover, Kessing et al.⁷⁸ showed anxiety level positively correlates to severity of heartburn and chest pain.

Hypervigilance is a mental condition which encompasses several thoughts that drives attention to body sensations, anxiety and expectation about symptoms and fear for the consequences of the symptoms. This condition lowers a threshold to stimuli, that is to say, the patients can perceive no or subtle stimuli which most people cannot (i.e., allodynia), and symptoms can be intensified (i.e., hyperalgesia).⁷⁹ Taft et al.⁷⁹ have recently developed and validated a questionnaire called esophageal hypervigilance and anxiety scale (EHAS) to assess hypervigilance and anxiety, specifically for esophageal sensations. So far several studies demonstrated close relationship between EHAS and esophageal symptom severity in GERD, esophageal motility disorders, and eosinophilic esophagitis.^80-83 Wong et al.⁸⁰ found that EHAS positively correlated to reflux symptom severity regardless of GERD phenotypes. While Guadagnoli et al.⁸¹ demonstrated that only hypervigilant component in EHAS was a significant factor associated with reflux symptom severity.

CONDITIONS ASSOCIATED WITH RH

1. IBS and function dyspepsia

Common functional gastrointestinal disorders such as FD and IBS often overlap with GERD.^84,85 Our previous study found that FD and IBS coexist with 31% and 29% of GERD patients respectively, and more importantly the overlap worsens health-related quality of life compared to having either one condition.⁸⁴ In a meta-analysis, IBS patients are more likely to have reflux symptoms compared to non-IBS patients (odds ratio, 4.17; 95% confidence interval, 2.85 to 6.09).⁸⁶ de Bortoli et al.⁸⁷ further scrutinized this relationship for each GERD phenotype and found that the prevalence of IBS in RH was 48.2% situating between true NERD (17%) and FH (77%). This tendency was similar in the prevalence of anxiety and depression diagnosis between the three groups. With regard to FD, Savarino et al.¹⁴ found that clinically relevant dyspeptic symptoms in 37% of RH as well as NERD patients, which was lower than FH (63%). These findings suggest that RH shares the mechanism of visceral hypersensitivity with other functional gastrointestinal disorders.

2. Behavioral disorders (supragastric belching and rumination syndrome)

Excessive supragastric belching (SGB) and rumination syndrome (RS) are functional gastroduodenal disorders in Rome IV.⁸⁸ Their impact on PPI refractory GERD has been increasingly recognized recently. In a study by Yadlapati et al.,⁸⁹ postprandial high-resolution impedance manometry revealed the high prevalence of excessive SGB (42%) and RS (20%) in 94 PPI nonresponders. Our study evaluating the prevalence of the behavioral disorders in each reflux phenotype found that approximately 40% and 10% of RH patients had excessive SGB and RS respectively. Of which, 40% of typical reflux episodes (i.e., heartburn and regurgitation) recorded in impedance-pH monitoring was associated with SGB or rumination events.⁹⁰

Belching can be classified into two types, gastric belching and SGB. Gastric belching is a physiological phenomenon to eliminate accumulated air in the proximal stomach through the mouth via transient lower esophageal relaxation. On the other hand, SGB is a behavioral disorder where a patient subconsciously swallows or sucks air from the mouth into the esophagus, then immediately followed by evacuating the air through the mouth using abdominal straining. SGB can cause reflux symptoms in two ways (1) triggering gastroesophageal reflux within a few seconds after the onset of SGB⁹¹ or (2) distending the esophagus. In fact, esophageal balloon distension can trigger heartburn sensation.⁹² Notably, some patients with excessive SGB predominantly complain about reflux symptoms rather than belching symptom.⁹¹

RS is defined as the repetitive, effortless regurgitation of recently ingested food into the mouth followed by rechewing and re-swallowing or expulsion of the food bolus.⁸⁸ Typical RS patients are young female with low body mass index suffering from regurgitation.⁹⁰ RS uses voluntary, but unconscious abdominal straining to bring up the gastric content into the mouth generating high intragastric pressure.⁹³

SGB and RS patients are often referred to a gastroenterologist as PPI refractory GERD since reflux symptoms due to these conditions do not respond to acid suppression. Therefore, careful medical interview and impedance-pH monitoring would be required to avoid overlooking them.

TREATMENT

1. Proton pump inhibitors

RH is often diagnosed after unsuccessful PPI/PCAB treatment. However, several studies evaluated the efficacy of PPI on PPI-naïve RH patients. Watson et al.⁹⁴ found that omeprazole (40 mg/day) relieved 67% of GERD patients with normal endoscopy and physiological acid exposure in which positive SI was related to better outcome. de Bortoli et al.⁸⁷ found that 56% of 218 RH patients experienced >50% symptom relief by PPIs, which is lower than NERD (68%), however, much better than FH (0%). Besides, Taghavi et al.⁹⁵ showed the good association between positive SAP and symptomatic response to omeprazole (60 mg/day). Given the high efficacy of high-dose PPIs, maximizing the acid suppression (e.g., double-dose PPIs or vonoprazan 20 mg/day) should be considered for RH patients naïve to such treatments. However, PPIs/PCAB should be stopped especially for hypersensitivity to nonacid reflux in off-PPI impedance-pH monitoring.²⁶

2. Histamine2-receptor antagonists

Histamine2-receptor antagonists are known to have analgesic effects on visceral nociception.⁹⁶ Rodriguez-Stanley et al.⁹⁶ showed ranitidine ameliorated esophageal acid sensitivity measured by acid perfusion test in RH or FH patients even after single administration of the drug. It is also the case in famotidine for NERD patients.⁹⁷ Due to the small sample size in these studies, the pain modulating effect of histamine2-receptor antagonists needs to be evaluated for a large RH cohort.

3. ARS (fundoplication)

There are some evidences to support ARS for RH patients. Broeders et al.⁹⁸ found that Nissen fundoplication improved or resolved reflux symptoms in 85% of RH patients even at 5 years after the surgery, which was similar to patients with pathological acid exposure. Patel et al.¹⁶ found the preferable outcome of ARS to medical therapy in RH. Although ARS achieved >50% symptoms improvement in about 90% of RH patients, the presence of hiatus hernia was associated with positive outcome. Besides, Rengarajan et al.⁹⁹ found that patients with normal AET could be a candidate for ARS when MNBI is abnormal (<2,292 Ω). It implicates that part of these RH patients might be NERD with false negative AET. Spechler et al.¹⁰⁰ conducted a randomized controlled study having three treatment arms (i.e., ARS [Nissen fundoplication], PPI plus baclofen and desipramine, and PPI plus placebo baclofen and desipramine) for refractory heartburn patients with positive SAP in on-PPI impedance-pH monitoring. They showed that the ARS group (67%) had better treatment success rate at 1 year after surgery than other groups, and the outcome in the ARS group did not differ regardless of coexisting abnormal acid exposure. Desjardin et al.¹⁰¹ also showed that positive SAP could predict successful outcome of ARS.

Taking these findings into consideration, RH is likely to be a good candidate for ARS as recommended in the ICARUS guidelines.¹⁰² Concomitant condition related to abnormal reflux (i.e., hiatus hernia or low MNBI) can further justify its application. However, there is still not enough evidence to support the indication. Furthermore, it should be noted that RH patients in several studies above are diagnosed by on-PPI impedance-pH monitoring. Given that PPIs/PCAB can normalize AET in most patients, RH should be diagnosed off PPI. Thus, we should be very careful about application of ARS to true RH phenotyped off PPI.

4. Pain modulators

Antidepressants are recommended as pain modulators for functional esophageal disorders. Several randomized controlled studies tested the effect of tricyclic antidepressants (TCA) and selective serotonin reuptake inhibitors (SSRIs) for well-phenotyped RH patients refractory to PPIs. Limsrivilai et al.¹⁰³ showed once-daily imipramine 25 mg (TCA) had no effect on reflux symptoms despite a marginal improvement of QOL. On the other hand, Viazis et al.¹⁰⁴ found that citalopram 20 mg/day (SSRI) resolved reflux symptoms in 61.5% of the RH patients superior to placebo (33.3%). Another study including both RH and FH patients also showed a better effect of fluoxetine (SSRI) over omeprazole and placebo.¹⁰⁵

5. Cognitive behavioral therapy

Excessive SGB and/or RS require dedicated intervention due to their unignorable impact on reflux symptom burden.⁹⁰ Although some studies reported the efficacy of baclofen and pregabalin on such behavioral disorders,^106,107 cognitive behavioral therapy using diaphragmatic breathing should be attempted first because of the equal or greater efficacy without any adverse effect than the drugs.^91,108-110 We reported that cognitive behavioral therapy reduces excessive SGB sufficiently in almost half of the patients.⁹¹ Our cognitive behavioral therapy program is composed of 5 sessions during 10 weeks. In brief, it helps patients to understand the mechanism of SGB in cognitive part and learn slow diaphragmatic breathing and mouth opening/tongue position to physically prevent SGB in behavioral part.^91,111 In RS, pain modulators (e.g., TCA) might be an additional option if refractory to the diaphragmatic breathing.¹¹²

CONCLUSION

Since the Montreal definition,¹¹³ our understanding of GERD has been extended by a great deal of studies. Especially impedance monitoring and prolonged wireless pH metry have contributed to better understanding of RH. This progress led to a new perspective of GERD proposed by Rome IV and the Lyon consensus so that more pathophysiology-focused diagnosis can be made for the four reflux phenotypes including RH. The discovery of fairly high impact of behavioral disorders on reflux symptoms would be another novel finding contributing to identification of true RH. More precise diagnosis is essential for better management. However, treatment modalities for esophageal hypersensitivity are still limited. Further study will be required to understand the pathophysiology of esophageal hypersensitivity and develop new treatments.