JNM

J Neurogastroenterol Motil, Vol. 29 No. 2 April, 2023

pISSN: 2093-0879 eISSN: 2093-0887
https://doi.org/10.5056/jnm22115
Journal of Neurogastroenterology and Motility Original Article

Prevalence of Non-erosive Esophageal Phenotypes

in Children: A European Multicenter Study

Elisa Blasi,1,8 Ettore Stefanelli,2 Renato Tambucci,3 Silvia Salvatore,4 Paola De Angelis,3 Paolo Quitadamo,5 Claudia Pacchiarotti,6
Giovanni Di Nardo,6 Fanj Crocco,7 Enrico Felici,7 Valentina Giorgio,8 Nicoletta Staropoli,9 Simona Sestito,2 Efstratios Saliakellis,1
Osvaldo Borrelli,1 and Licia Pensabene1,2*
1
Neurogastroenterology and Motility Unit, Department of Gastroenterology, Great Ormond Street Hospital for Children, London, UK;
2
Department of Medical and Surgical Sciences, Pediatric Unit, University Magna Graecia of Catanzaro, Catanzaro, Italy; 3Digestive Endoscopy
and Surgery Unit, Bambino Gesù Children's Hospital - Research Institute IRCCS, Rome, Italy; 4 Pediatric Department, University of Insubria,
Hospital “F. Del Ponte”, Varese, Italy; 5Department of Pediatrics, A.O.R.N. Santobono-Pausilipon, Napoli, Italy; 6NESMOS Department, Faculty
of Medicine and Psychology, Sapienza University of Rome, Pediatric Unit, Sant’Andrea University Hospital, Rome, Italy; 7Pediatric and Pediatric
Emergency Unit, Children Hospital, AO SS Antonio e Biagio e C. Arrigo, Alessandria, Italy; 8Department of Woman and Child Health and Public
Health, Public Health Area, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy; and 9A.O.U. Mater Domini, University Magna
Graecia of Catanzaro, Catanzaro, Italy

Background/Aims
Since available data on pediatric non-erosive esophageal phenotypes (NEEPs) are scant, we investigated their prevalence and the
phenotype-dependent treatment response in these children.
Methods
Over a 5-year period, children with negative upper endoscopy, who underwent esophageal pH-impedance (off-therapy) for persisting
symptoms not responsive to proton pump inhibitor (PPI)-treatment, were recruited. Based on the results of acid reflux index (RI) and
symptom association probability (SAP), patients were categorized into: (1) abnormal RI (non-erosive reflux disease [NERD]), (2) normal
RI and abnormal SAP (reflux hypersensitivity [RH]), (3) normal RI and normal SAP (functional heartburn [FH]), and (4) normal RI and
not-reliable SAP (normal-RI-not otherwise-specified [normal-RI-NOS]). For each subgroup, treatment response was evaluated.
Results
Out of 2333 children who underwent esophageal pH-impedance, 68 cases, including 18 NERD, 14 RH, 26 FH, and 10 normal-RI-NOS
were identified as fulfilling the inclusion criteria and were analyzed. Considering symptoms before endoscopy, chest pain was more
reported in NERD than in other cases (6/18 vs 5/50, P = 0.031). At long-term follow-up of 23 patients (8 NERD, 8 FH, 2 RH, and 5
normal-RI-NOS): 17 were on PPIs and 2 combined alginate, 1 (FH) was on benzodiazepine + anticholinergic, 1 (normal-RI-NOS) on
citalopram, and 3 had no therapy. A complete symptom-resolution was observed in 5/8 NERD, in 2/8 FH, and in 2/5 normal-RI-NOS.
Conclusions
FH may be the most common pediatric NEEP. At long-term follow-up, there was a trend toward a more frequent complete symptom
resolution with PPI-therapy in NERD patients while other groups did not benefit from extended acid-suppressive-treatment.
(J Neurogastroenterol Motil 2023;29:156-165)

Key Words
Children; Functional heartburn; Gastroesophageal reflux disease; Non-erosive reflux disease; Reflux hypersensitivity

Received: July 15, 2022 Revised: January 24, 2023 Accepted: January 24, 2023
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.
org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work
is properly cited.
*Correspondence: Licia Pensabene, MD, PhD
Department of Medical and Surgical Sciences, Pediatric Unit, University Magna Graecia of Catanzaro, Viale Europa, Località
Germaneto, 88100 Catanzaro, Italy
Tel: +39-0961-3647777, E-mail: pensabene@unicz.it
Osvaldo Borrelli and Licia Pensabene act equally as senior authors.

ⓒ 2023 The Korean Society of Neurogastroenterology and Motility

156 J Neurogastroenterol Motil, Vol. 29 No. 2 April, 2023

www.jnmjournal.org
 Non-erosive Esophageal Phenotypes in Children

it may have a less likely response to acid-suppressive-treatment be-

cause of normal esophageal acid exposure.
Introduction So far, only a single retrospective pediatric study involving 45
Gastroesophageal reflux disease (GERD), defined as bother- children has assessed the prevalence of non-erosive reflux disease
some symptoms and/or complications resulting from gastric con- subgroups in pediatric ages.15 In this European multicenter study,
tents reflux into the esophagus or beyond,1,2 is a chronic condition we aim at investigating the prevalence of the Rome IV NEEPs
characterized by a heterogeneous spectrum of clinical manifesta- (NERD, RH, and FH), and of the subgroup with normal esopha-
tions.3 Visible endoscopic breaks/erosions of the esophageal mucosa geal acid exposure and a not-reliable symptom association (undeter-
detected on esophago-gastro-endoscopy identify patients with mined phenotype), in a cohort of children who underwent both en-
erosive reflux disease (ERD).3 Conversely, the presence of micro- doscopy and pH-impedance off-therapy. Moreover, the long-term
scopic esophagitis without evidence of erosive esophagitis is not outcome of symptoms and the response to treatment were examined
considered pathognomonic of GERD as it can also be found in up in each subgroup.
to 15% of asymptomatic healthy controls.4 Nonetheless, an increas-
ing body of evidence has shown that,5-7 in a variable proportion of
patients, GERD may exist in the absence of erosive esophagitis, as
Materials and Methods
demonstrated by the presence of pathologic acid reflux on esopha- The medical records of all children, aged 5-17 years, between
geal pH-(impedance) monitoring.6,7 In this case the phenotype January 2014 and April 2019, who underwent MII-pH within 6
is defined as non-erosive reflux disease (NERD) and a differ- months from a normal endoscopy for persisting typical reflux symp-
ent treatment response compared to ERD has been reported.8-10 toms (such as heartburn or epigastric pain) despite acid suppression
Whether endoscopy should be performed while the patient is on treatment (4 weeks), were retrospectively reviewed. The list of pa-
or off acid suppression is still debated. Adult guidelines on GERD tients was obtained by the existing MII-pH electronic database of
suggest performing endoscopy off acid suppression therapy to allow the 6 participating pediatric gastroenterology centers. Children were
a correct identification of the type of esophagitis.11,12 The current included if they underwent MII-pH off-therapy, within 6 months
European and Nord American guidelines on GERD3 recom- from a normal endoscopic assessment also performed off-therapy;
mended performing a combined 24-hour multichannel intraluminal therefore both investigations had to be performed off-therapy, that
impedance and pH monitoring (MII-pH) in those children with meant a wash-out period of at least 4 weeks before endoscopy, if
normal endoscopy but persistent symptoms despite acid suppres- previously treated with acid suppressive drugs. Before being sub-
sion therapy for 4-8 weeks. Based on the results of this diagnostic mitted to MII-pH, patients should be off at least 2 weeks for acid
testing, the Rome IV criteria3 on esophageal disorder define 3 dis- suppressants, 72 hours if treated with prokinetics, and 4 hours if
tinct non-erosive esophageal phenotypes (NEEPs): (1) those with they had alginate. We considered a normal endoscopy in the absence
abnormal esophageal acid exposure regardless of symptom cor- of macroscopically visible breaks/erosions in the esophageal mucosa.
relation (NERD), (2) those with normal esophageal acid exposure The presence of microscopic esophagitis, defined as inflammatory
but a positive symptom association to acid or nonacid reflux (reflux cell infiltration, basal zone hyperplasia, papillary elongation, dilata-
hypersensitivity, RH), and (3) those with normal esophageal acid tion of intercellular spaces,16 and evaluated in all patients by endo-
exposure and a negative symptom association (functional heartburn, scopic biopsies, was not an exclusion criterion. Exclusion criteria
FH). were represented by one of the following: erosive or eosinophilic
Appropriate categorization of NEEP patients has important esophagitis; past history of gastric, duodenal or esophageal surgery;
therapeutic implications since each subgroup has different patho- known esophageal motor disorders or any condition that interferes
genic mechanisms and may respond differently to medical and with the absorption, distribution, and metabolism of drugs (eg,
surgical interventions.13,14 Moreover, in clinical practice, physicians celiac, inflammatory bowel disease); systemic disease (eg, diabetes
may face with a fourth subgroup of patients with normal esophageal mellitus, peripheral and autonomic neuropathies); neurological or
acid exposure and a not-reliable symptom association, because of mental impairment; major depression or behavioral disorder; drug
a very limited (less than 3) symptom episodes reported during the or alcohol abuse; children on therapy with neuroleptic agents or any
investigation characterization being an uncertain phenotype. Al- antidepressant during 6 months prior to enrollment; and children
though this subgroup has not been defined by the Rome IV criteria, not fulfilling the inclusion criteria.

Vol. 29, No. 2 April, 2023 (156-165) 157

Elisa Blasi, et al

Collected data from medical records included demographic, occurring at the time of MII-pH.
reported symptoms before and after diagnostic testing, macroscopic Informed consent was obtained by parents of recruited chil-
and histologic endoscopic findings, and MII-pH results. dren. The study was approved by the Ethic Committees (No. 256,
MII-pH was performed and analyzed in each hospital as 19th September, 2019).
previously described.3,17-20 The MII-pH collected data included:
number, characteristics and proximal extension of reflux episodes,
reflux index (RI), number and type of symptoms reported during
Statistical Methods
recording, symptom association probability (SAP), post reflux swal- Data are presented as percentage (%), mean ± SD, median
low induced peristaltic wave (PSPW), and mean nocturnal baseline with interquartile range, as appropriate. The normality of distribu-
impedance (MNBI). tion for continuous variables was assessed with Shapiro-Wilk test.
The reflux index (RI), defined as the total percent time of acid Continuous variables were analyzed with Student’s t test or Wilcox-
exposure, was considered abnormal if pH < 4 for > 5% of the on signed rank test as appropriate. Comparisons between categori-
study duration.17-20 cal variables were performed with either chi-square or Fisher’s exact
The symptom association probability (SAP) index, currently test as appropriate. Differences in the mean values (continuous vari-
recognized as the most accurate statistical parameter for reflux- ables) among disease phenotype groups were assessed with one-way
symptom association analysis,21 was considered positive when > ANOVA or Kruskal-Wallis test as appropriate. The adjusted effect
95%. The minimum number of symptoms to produce a reliable of various factors on certain outcomes of interest were explored
analysis is still debated and is related to the type of symptoms and with the appropriate type of regression analysis. All P -values were 2
the method of recording.22 Nevertheless, it is widely recognized that sided with P less than 0.05 being considered statistically significant.
the fewer the number of symptoms reported, the less reliable the
SAP becomes.23 Thus, in our study we considered as not-reliable
SAP when the symptom was reported less than 3 times during the
Results
MII-pH. Based on MII-pH results, we categorized the enrolled During the 5-year period, 2333 MII-pH impedances were
patients in 4 subgroups: (1) abnormal RI (NERD); (2) normal RI performed in the participating pediatric centers and 68 children
and abnormal SAP (RH); (3) normal RI and normal SAP (FH); (median age 11 ± 3.5 years, 33 female) were identified as fulfilling
and (4) normal RI and not-reliable SAP (normal RI not otherwise the study criteria and were included in the analysis. Based on RI
specified [“normal RI-NOS”]). and SAP, patients were categorized into: NERD 18/68 (26.5%),
The post reflux swallow induced peristaltic wave (PSPW) RH 14/68 (20.6%), FH 26/68 (38.2%), and normal RI-not oth-
was defined as an antegrade 50% drop in impedance starting in erwise-specified (RI-NOS) 10/68 (14.7%) (Table 1). The preva-
the most proximal impedance channel and reaching the most dis- lence of symptoms pre-endoscopy in the study population and the
tal impedance channel, occurring within 30 seconds after a reflux distribution of each symptom among different NEEPs are shown
event.24 The PSPW index was then obtained dividing the number in Figure. Regarding pre-endoscopy symptoms and phenotypes,
of PSPWs by the total number of reflux events.25 chest pain was significantly more frequent in NERD (6/18) and
MNBI was assessed from the most distal impedance chan- in normal RI-NOS (4/10), (P = 0.020), whilst chest pain was re-
nel during night-time by calculating the mean impedance baseline ported in only 1/26 FH children and in none of the 14 RH patients
among three 10-minute time periods (at 1, 2, and 3 AM) with no (Figure). When assessing the association between symptoms pre-
reflux episodes, pH drops or swallows.26 According to previous endoscopy and gender, we found that heartburn was more frequent
published adult criteria, cutoff values for PSPW index and MNBI in females (22/33 vs 9/35, P = 0.001), while there was a trend not
were 61% and 2292 Ω, respectively.27 reaching statistical significance toward an increase complaint of ab-
Finally, a telephone interview was conducted to evaluate, in each dominal pain in male (15/35 vs 8/33, P = 0.120).
subgroup, symptoms persistence 2 to 5 years after investigations. No significant statistical difference was found when comparing
Parents or patients (when older than 12 years) were asked about the phenotypes and gender, histology report, and persisting symptoms
presence, partial (> 50% decrease in frequency and/or intensity of during treatment.
symptoms) or complete symptoms resolution on or off treatment in The prevalence of persisting symptoms during proton pump
the month before the recall, and in comparison with the symptoms inhibitor (PPI) therapy among the different phenotypes is shown

158 Journal of Neurogastroenterology and Motility

 Non-erosive Esophageal Phenotypes in Children

Table 1. Demographic Characteristics Among Subtypes

NERD RH FH nl-RI-NOS Total P -value

Total n (%) 18 (26.5) 14 (20.6) 26 (38.2) 10 (14.7) 68 (100) NS
Male n (%) 10 (55.5) 7 (50) 14 (53.8) 4 (40) 35 (51.5) NS
Median age (IQR) 12.5 (8-14) 10.5 (8-12) 11 (6-13) 11 (9-16) 11 (8-13) NS
NERD, non-erosive reflux disease; RH, reflux hypersensitivity; FH, functional heartburn; nl-RI-NOS, normal reflux index not otherwise-specified; IQR, inter-
quartile range; NS, not significant.

in Figure. There was a trend without statistical significance toward subgroups and according to age.
an increased report of regurgitation, abdominal pain, and heartburn
compared to the other persisting symptoms and an increased report Telephone Interview
of vomiting and chest pain in NERD patients. At recall (mean ± SD follow-up duration: 28.8 ± 21.8
The prevalence of reported symptoms during MII-pH in the months) data of 23 patients (10 female, 8 NERD, 8 FH, 2 RH,
entire study population and the distribution among different pheno- and 5 normal RI-NOS) were available. Among them, 17 were
types is illustrated in Figure. treated with PPIs: 13 only PPI (6 with NERD), 2 combined with
magnesium alginate, 1 was also on benzodiazepine plus anti-cholin-
Endoscopy and Biopsy ergic drug, and 1 on citalopram; 3 patients were treated only with
Microscopic esophagitis was found in 28.6% of patients: in magnesium alginate (1 FH, 1 RH, and 1 normal RI-NOS) (Table
33.3% (5/15) of patients with NERD, in 28.6% (4/14) of patients 3). In the last month of follow up, there was no need of therapy in
with RH, in 23.1% (6/26) with FH and in 37.5% (3/8) with nor- 13 patients (4 NERD, 4 FH, and 5 normal-RI-NOS) and the
mal RI-NOS; in the remaining 5 patients (3 with NERD and 2 need of a course of treatment in 3 NERD and in 4 FH patients,
with normal RI-NOS) the referral letter from outside hospital ge- while 1 RH patients was treated as needed. A complete symptom
nerically reported “normal histology” without a detailed histological resolution was observed in 9/23 (39.0%) children: 5/8 (62.5%)
description, therefore these 5 patients were not included in this sub- NERD, in 2/8 (25.0%) FH (1 treated also with benzodiazepine
analysis. No significant differences in distribution of microscopic plus anticholinergic), and in 2/5 (40.0%) normal RI-NOS (1 treat-
esophagitis among groups were found. ed with the addition of citalopram). There was a partial symptom
resolution in 3/8 (37.5%) NERD and in 6/8 (75.0%) FH, in 1/2
Multichannel Intraluminal Impedance and pH (50.0%) RH and in 1/5 (20.0%) normal-RI-NOS (Table 3).
Monitoring
The MII-pH results are shown in Table 2. NERD patients
presented a higher median number of acid and total reflux episodes
Discussion
compared to the other groups and when comparing NERD to To the best of our knowledge, this is the largest described
FH (respectively P = 0.001 and P = 0.008). Moreover, NERD cohort of children with NEEPs based on the Rome IV-criteria.
patients presented a higher median duration of the longest reflux Increasing evidence suggest the existence of a phenotypic spectrum
episodes, statistically significant when comparing NERD to FH of GERD, with multifactorial underlying mechanisms, leading dif-
(P = 0.004), and a higher number of proximal reflux episodes, ferent symptom perception and possible treatment response.28 The
comparing NERD to normal RI-NOS (P = 0.001). MNBI most common presentation of GERD at any age is non-erosive
values were lower in NERD and in normal RI-NOS (statistically reflux disease, which has recently characterized in 3 distinct pheno-
significant only between normal RI-NOS and RH and FH) (Table types (ie, NERD, FH, and RH) by the Rome IV esophageal cri-
2). PSPW index was lower in NERD children compared to other teria.3,14 The prevalence of erosive esophagitis is even lower in chil-
phenotypes with no statistical difference among NEEPs (Table 2). dren compared to adults.29,30 So far, only a single pediatric study15
In the NERD group, 33.3% (6/18) of patients had a positive SAP. examined the NEEPs prevalence by using the Rome IV criteria:
RH had positive SAP only for acid reflux-associated symptoms. among 45 children aged ≥ 5 years who underwent both endoscopy
Regarding the reported symptoms during pH-MII (Figure), we and pH-MII testing off PPI-therapy for typical gastroesophageal
did not found differences in the distribution of symptoms among reflux symptoms, 44.0% were diagnosed as having FH, 29.0% as

Vol. 29, No. 2 April, 2023 (156-165) 159

Elisa Blasi, et al

Symptoms before endoscopy

Vomiting Heartburn Abdominal pain Belching
6 NERD 7 NERD 5 NERD 2 NERD
2 RH 4 RH 7 RH 2 RH
6 FH 14 FH 9 FH 5 FH
1 nl-RI-NOS 6 nl-RI-NOS 2 nl-RI-NOS

Total = 15 Total = 31 Total = 23 Total = 9

Regurgitation Nausea Cough Cheat pain
5 NERD 2 NERD 5 NERD 6 NERD*
6 RH 4 RH 3 RH 1 FH
6 FH 5 FH 7 FH 4 nl-RI-NOS
3 nl-RI-NOS 1 nl-RI-NOS 1 nl-RI-NOS
*P = 0.020

Total = 20 Total = 12 Total = 16 Total = 11

Symptoms on PPI
Vomiting Heartburn Abdominal pain Belching
5 NERD 3 NERD 3 NERD 1 NERD
1 RH 2 RH 5 RH 2 RH
1 FH 7 FH 7 FH 3 FH
2 nl-RI-NOS 2 nl-RI-NOS

Total = 7 Total = 14 Total = 17 Total = 6

Regurgitation Nausea Cough Cheat pain
4 NERD 2 NERD 3 NERD 4 NERD
5 RH 3 RH 2 RH 1 FH
6 FH 3 FH 1 FH 1 nl-RI-NOS
2 nl-RI-NOS

Total = 17 Total = 8 Total = 6 Total = 6

Symptoms during MII-pH

Vomiting Heartburn Abdominal pain Belching
3 NERD 2 NERD 3 NERD 1 NERD
2 RH 3 RH 4 RH 2 RH
1 FH 8 FH 3 FH 3 FH

Total = 6 Total = 13 Total = 10 Total = 6

Regurgitation Nausea Cough Cheat pain
4 NERD 1 NERD 3 NERD 2 NERD
8 RH 3 RH 3 RH 1 nl-RI-NOS
7 FH 2 FH 14 FH
4 nl-RI-NOS 1 nl-RI-NOS

Total = 23 Total = 6 Total = 21 Total = 6

Figure. Prevalence of symptoms and distribution among different phenotypes. NERD, non-erosive reflux disease; RH, reflux hypersensitivity;
FH, functional heartburn; nl-RI-NOS, normal reflux index not otherwise specified; PPI, proton pump inhibitor; MII-pH, multichannel intralu-
minal impedance and pH monitoring.

160 Journal of Neurogastroenterology and Motility

 Non-erosive Esophageal Phenotypes in Children

Table 2. Multichannel Intraluminal Impedance and pH Monitoring Parameters

NERD RH FH nl-RI -NOS P -valuea

Esophageal acid exposure time (RI) 11.4 (8.3-14.7) 2.2 (0.5-3.3) 0.7 (0.2-1.4) 1.5 (0.8-3.6) < 0.001b,c,d
(median, IQR)
Total reflux number (median, IQR) 87 (35.3-100) 41.5 (41.5-75.8) 28 (19.3-39.3) 24 (9.7-47) 0.008b
pH only events (median, IQR) 104 (81-121) 25.5 (10-42.8) 15 (3.7-25.3) 36 (9-57) < 0.001b,c
Acid reflux number (median, IQR) 63 (33.3-80.8) 27.5 (10.3-41.3) 14.5 (4.2-26) 16 (3-32) 0.001b
Duration longest reflux (median, IQR) 20 (9.2-24) 2.6 (1.3-3.9) 1.9 (1.1-2.8) 7.9 (1.7-7.9) 0.004b
Proximal reflux episodes number 40 (25-96) 32 (17.5-60.3) 14 (8-22) 8 (2.5-24.4) < 0.001b,d
(median, IQR)
Distal MNBI (median, IQR) 1315 (1018-2832) 2724 (2273-3403) 2576 (2115-3014) 1446 (1165-1749) 0.003e,f
PSPW % (median, IQR) 42.6 (29.6-45.8) 56.3 (38.7-67.2) 52 (35.9-69.1) 59 (43.3-81.5) 0.125
a
Comparisons with non-parametric ANOVA (Kruskal-Wallis test with Dwass-Steel-Critchlow-Fligner pairwise comparisons).
Levels of statistical significance: bNERD – FH, cNERD – RH, dNERD – RI neg-NOS, eFH - RI neg-NOS, and fRH - RI neg-NOS.
NERD, non-erosive reflux disease; RH, reflux hypersensitivity; FH, functional heartburn; nl-RI-NOS, normal reflux index not otherwise-specified; RI, reflux in-
dex; IQR, interquartile range; MNBI, mean nocturnal baseline impedance; PSPW, post reflux swallow induced peristaltic wave.

Table 3. Data at Follow-up

NERD RH FH nl-RI-NOS Total

Total n (%) 8 (34.8) 2 (8.7) 8 (34.8) 5 (21.7) 23 (100.0)
Male n (%) 5 (62.5) 1 (50.0) 4 (50.0) 3 (60.0) 13 (56.5)
Treatment after MII-pH (n) NERD RH FH nl-RI-NOS Total
PPI therapy and/or magnesium alginate 7 2 5 4 18
PPI and neuromodulators 0 0 1 1 2
No therapy 1 0 2 0 3
Treatment response at follow-up (n [%]) NERD RH FH nl-RI-NOS Total
Disappearance of symptoms: 5 (62.5) - 2 (25) 2 (40) 9 (39.1)
Reduction of > 50% of symptoms: 3 (37.5) 1 (50) 6 (75) 1 (20) 11 (47.8)
Persistence of symptoms: - 1 (50) - 2 (40) 3 (13.1)
NERD, non-erosive reflux disease; RH, reflux hypersensitivity; FH, functional heartburn; nl-RI-NOS, normal reflux index not otherwise-specified; MII-pH,
multichannel intraluminal impedance and pH monitoring; PPI, proton pump inhibitor.

RH and 27.0% as NERD, with the latter more commonly diag- Several different underlying mechanisms have been sug-
nosed in older children.15 gested to explain the discrepancy in non-erosive GERD phenotype
Our multicenter study confirms, in a larger cohort of European prevalence between children and adults.32 Mahoney et al15,28 hy-
children, that FH is the most common pediatric NEEP (38.2%,), pothesized a predominance of peripheral and central sensitization
followed by NERD (26.5%) and acid RH (20.6%). It is worth to in children.33,34 In an adult study, patient with FH have shown a
note that in 14.7% of children with normal RI, SAP was not reli- significantly higher balloon distention mechano-sensitivity and acid
able because of almost absence of symptoms (< 3 episodes) during perfusion chemosensitivity, when compared to either patients with
MII-pH; in this subgroup the phenotype was underdetermined NERD or healthy subjects.35 Several genetic risk factors related to
(normal-RI-NOS). Nevertheless, even categorizing to one extreme pain and molecular biomarkers have also been reported in associa-
all children with normal-RI-NOS as possible RH, the most com- tion with increased symptom perception.36-38 Moreover, in adult
mon NEEP in our cohort would remain FH. patients with FH, the afferent nerve fibers in the distal esophagus
In contrast, studies conducted in adults found a higher preva- were distributed similarly to healthy asymptomatic controls, and
lence (35.0-52.0%) of NERD phenotype, RH ranged from 14.0% both groups had significantly deeper nerve fibers in the mucosa,
to 35.0% and FH was identified in 22.0-30.0% of patients.5,7,31 away from the luminal surface, compared to patients with NERD,39

Vol. 29, No. 2 April, 2023 (156-165) 161

Elisa Blasi, et al

supporting the hypothesis that heartburn in FH may have a dis- any other specific symptom profile that could be useful in prediction
tinct nociceptive pathophysiology. In contrast, Woodland et al40 of reflux phenotype. In 62 adult patients, Kandulski et al55 found no
found significantly more superficial esophageal afferent nerves differences in reported symptoms (heartburn, regurgitation, or dys-
in adult NERD patients as compared to ERD patients, Barrett pepsia) between NERD, erosive esophagitis, and FH. In contrast,
esophagus, and controls. Very recently,41 it has been reported that Savarino et al56 showed a higher prevalence of heartburn in FH and
in NERD children the esophageal mucosa displays deep lying of epigastric pain in NERD patients.
nerve fibers and do not express the acid-sensing transient receptor As for the symptom frequency and association during MII-
potential channel vanilloid type 1 (TRPV1), in contrast to adults pH, we did not find significant differences among the phenotypes.
with NERD,40 who showed a more superficial esophageal mucosa We noticed a trend of increased episodes of vomiting and chest pain
innervation and a TRPV1 overexpression. Different age-dependent in NERD and the association limited to acid reflux in RH, likely
levels of inflammation and/or repair mechanisms have also been due to the scarcity of non-acid reflux episodes in this age group, as
considered.41 The absence of severe inflammatory infiltration of the also shown by Mahoney et al.15
esophageal mucosa may explain the deep position of the afferent Regarding histological features, overall, 28.6% patients present-
nerve fibers in children with NERD, since basal cell hyperplasia ed microscopic esophagitis, with a trend without statistical signifi-
and papillary elongation would not move nerve fibers towards the cance toward a more frequent identification in NERD and normal-
luminal surface. In support of this hypothesis, there is a marked RI-NOS. A correlation with NERD phenotype was reported in
difference between the overall incidence of microscopic esophagitis adult patients55,57 but not in the previous pediatric study.15
in pediatric NERD (20.0%)15 (28.6% in our cohort) compared to In our population a lower distal MNBI was noted in NERD
adult NERD (76.0%).42 Moreover, impaired repair mechanisms and normal-RI-NOS (median 1315 and 1446 respectively) com-
in adults may also contribute toward the higher incidence of erosive pared to RH and FH. This result aligns with the presence of
esophagitis, 30.0% in adults43 vs 12.4% in children.29 Considering (microscopic) inflammation, reflux-induced impairment of mucosal
all these findings, NERD children seem to have less esophageal integrity and acid exposure in NERD.26,58-60 PSPW index was
inflammation and deeper nerve fiber position compared to adults. reported as able to discriminate between GERD and non-GERD
Esophageal hypervigilance, a form of hyperawareness and early adult subjects as well as NERD from FH.61 We found a lower
detection of painful esophageal stimuli, is independently and sig- PSPW in NERD children compared to other phenotypes with no
nificantly associated with symptoms severity and is consistent across statistical difference among NEEPs.
reflux groups.32 Nonetheless, in patients with FH, visceral neural Data on at least 2 years of follow-up were available for 23 chil-
pathways dysfunction and/or cortical processing alterations may also dren and 39.0% of cases reported a complete resolution of symp-
contribute and mediate esophageal hypersensitivity.44 Interestingly, toms at recall, with a higher percentage of NERD patients respon-
evoked cortical responses are produced by mucosal acid exposure in sive to PPI therapy (62.5%) compared to the other groups (25.0%
FH subjects prior to inducing heartburn.45 Moreover, there is also in FH). Interestingly, 1 patient with FH was treated successfully
a role for brain-gut interplay in symptom perception. Patients with with an association of benzodiazepine and anticholinergic, and 1
FH more often report other functional gastrointestinal disorders, patient of the normal-RI-NOS group improved with the addition
exhibit psychological comorbidity and somatization compared to of a neuromodulator (citalopram). Thus, extending PPI therapy
heathy volunteers and NERD.46-48 Stress has also been found to seems not appropriate for subgroups other than NERD patients
influence pain perception to esophageal stimuli.49-51 FH is associ- and acidic RH, while other phenotypes may benefit from different
ated with significant psychosocial distress, anxiety, depression, and treatment. In particular, in patients with normal-RI-NOS, extend-
impaired quality of life.52-54 However, if these underlying patho- ing PPI therapy seems not appropriate for unresponsive patients
genic mechanisms of FH and comorbidity may be more relevant and different treatment such as neuromodulators may be indicated.
in inducing symptoms perception in children as compared to adults The enrolling criteria, accountable for the small sample size
remains a matter of research. of our study, together with the retrospective design and the limited
Regarding the correlation between pre-endoscopy symptoms number of cases at follow-up, are significant pitfalls of our study,
and phenotypes, we found that chest pain was reported by 33.3% of limiting the possibility to draw definitive conclusion, especially on
NERD patients and in 4/10 of children with normal-RI-NOS but treatment efficacy. However, by restricting the analysis to children
not in patients with RH. In line with Mahoney,15 we did not find who underwent both endoscopy and MII-pH off acid suppression

162 Journal of Neurogastroenterology and Motility

 Non-erosive Esophageal Phenotypes in Children

treatment, we were able to properly characterize the NEEPs and 5. Cheng FK, Albert DM, Maydonovitch CL, Wong RK, Moawad
avoid the inclusion of possible other disorders such eosinophilic FJ. Categorization of patients with reflux symptoms referred for pH
and impedance testing while off therapy. Clin Gastroenterol Hepatol
esophagitis.
2015;13:867-873.
In conclusion, we confirmed the existence of different NEEPs 6. Modlin IM, Hunt RH, Malfertheiner P, et al. Diagnosis and manage-
in children, with the most common phenotype being FH. A com- ment of non-erosive reflux disease—the vevey NERD consensus group.
plete symptom resolution with PPI therapy occurs more frequently Digestion 2009;80:74-88.
in NERD patients while other phenotypes do not benefit from 7. Savarino E, Zentilin P, Savarino V. NERD: an umbrella term includ-
extended acid suppressive treatment. Further prospective studies ing heterogeneous subpopulations. Nat Rev Gastroenterol Hepatol
2013;10:371-380.
are needed to confirm these observations and to identify the best
8. Dean BB, Gano AD Jr, Knight K, Ofman JJ, Fass R. Effectiveness of
targeted therapeutic approach for each NEEP (eg, neuromodula- proton pump inhibitors in nonerosive reflux disease. Clin Gastroenterol
tors, complementary therapies), which still remains an unmet clini- Hepatol 2004;2:656-664.
cal need. 9. Fiedorek S, Tolia V, Gold BD, et al. Efficacy and safety of lansoprazole in
adolescents with symptomatic erosive and non-erosive gastroesophageal
Financial support: None. reflux disease. J Pediatr Gastroenterol Nutr 2005;40:319-327.
10. Tolia V, Ferry G, Gunasekaran T, Huang B, Keith R, Book L. Efficacy
Conflicts of interest: None. of lansoprazole in the treatment of gastroesophageal reflux disease in chil-
dren. J Pediatr Gastroenterol Nutr 2002;35(suppl 4):S308-S318.
Author contributions: Elisa Blasi: preparation of synopsis 11. Katz PO, Gerson LB, Vela MF. Guidelines for the diagnosis and
data, analysis of results, drafting the manuscript, and approval of management of gastroesophageal reflux disease. Am J Gastroenterol
the final version of the paper; Ettore Stefanelli, Renato Tambucci, 2013;108:308-328.
Silvia Salvatore, Paola De Angelis, Paolo Quitadamo, Claudia Pac- 12. Pouw RE, Bredenoord AJ. Mistakes in the use of PPIs and how to avoid
chiarotti, Giovanni Di Nardo, Fanj Crocco, Enrico Felici, Valentina them. UEG Educ 2017;17:15-17.
13. Yadlapati R, Pandolfino JE. Personalized approach in the work-up and
Giorgio, and Simona Sestito: preparation of synopsis data, critical
management of gastroesophageal reflux disease. Gastrointest Endosc
revision of the manuscript, and approval of the final version of the
Clin N Am 2020;30:227-238.
paper; Nicoletta Staropoli and Efstratios Saliakellis: analysis and in- 14. Aziz Q, Fass R, Gyawali CP, Miwa H, Pandolfino JE, Zerbib F. Func-
terpretation of data, critical revision of the manuscript, and approval tional esophageal disorders. Gastroenterology 2016;150:13681379.
of the final version of the paper; and Osvaldo Borrelli and Licia 15. Mahoney LB, Nurko S, Rosen R. The prevalence of rome IV nonerosive
Pensabene: conception and study design, analysis and interpretation esophageal phenotypes in children. J Pediatr 2017;189:86-91.
16. Yerian L, Fiocca R, Mastracci L, et al. Refinement and reproducibility
of data, critical revision of the manuscript, and approval of the final
of histologic criteria for the assessment of microscopic lesions in patients
version of the paper.
with gastroesophageal reflux disease: the esohisto project. Dig Dis Sci
2011;56:2656-2665.
17. Wenzl TG, Benninga MA, Loots CM, Salvatore S, Vandenplas Y;
References ESPGHAN EURO-PIG Working Group. Indications, methodology,
1. Vakil N, van Zanten SV, Kahrilas P, Dent J, Jones R; Global Consensus and interpretation of combined esophageal impedance-pH monitoring in
Group. The Montreal definition and classification of gastroesophageal children: ESPGHAN EURO-PIG standard protocol. J Pediatr Gas-
reflux disease: a global evidence-based consensus. Am J Gastroenterol troenterol Nutr 2012;55:230-234.
2006;101:1900-1920. 18. Mutalib M, Rawat D, Lindley K, et al. BSPGHAN motility working
2. Sherman PM, Hassall E, Fagundes-Neto U, et al. A global, evidence- group position statement: paediatric multichannel intraluminal pH im-
based consensus on the definition of gastroesophageal reflux disease in the pedance monitoring- indications, methods and interpretation. Frontline
pediatric population. Am J Gastroenterol 2009;104:1278-1295. Gastroenterol 2017;8:156-162.
3. Rosen R, Vandenplas Y, Singendonket M, et al. Pediatric gastroesopha- 19. Pilic D, Fröhlich T, Nöh F, et al. Detection of gastroesophageal reflux in
geal reflux clinical practice guidelines: joint recommendations of the north children using combined multichannel intraluminal impedance and pH
american society pediatric gastroenterology, hepatology and nytrition and measurement: data from the German pediatric impedance group. J Pedi-
the european society for pediatric gastroenterology, hepatology, and nutri- atr 2011;158:650-654, e1.
tion. J Pediatr Gastroenterol Nutr 2018;66:516-554. 20. Quitadamo P, Tambucci R, Mancini V, et al. Esophageal pH-impedance
4. Winter HS, Madara JL, Stafford RJ, Grand RJ, Quinlan JE, Gold- monitoring in children: position paper on indications, methodology
man H. Intraepithelial eosinophils: a new diagnostic criterion for reflux and interpretation by the SIGENP working group. Dig Liver Dis
esophagitis. Gastroenterology 1982;83:818-823. 2019;51:1522-1536.

Vol. 29, No. 2 April, 2023 (156-165) 163

Elisa Blasi, et al

21. Weusten BL, Roelofs JM, Akkermans LM, Van Berge-Henegouwen 37. Guarino MP, Cheng L, Ma J, et al. Increased TRPV1 gene expression
GP, Smout AJ. The symptom-association probability: an improved in esophageal mucosa of patients with non-erosive and erosive reflux dis-
method for symptom analysis of 24-hour esophageal pH data. Gastroen- ease. Neurogastroenterol Motil 2010;22:746-751, e219.
terology 1994;107:1741-1745. 38. Yoshida N, Kuroda M, Suzuki T, et al. Role of nociceptors/neuropeptides
22. Omari TI, Schwarzer A, vanWijk MP, et al. Optimisation of the reflux- in the pathogenesis of visceral hypersensitivity of nonerosive reflux dis-
symptom association statistics for use in infants being investigated by 24- ease. Dig Dis Sci 2013;58:2237-2243.
hour pH impedance. J Pediatr Gastroenterol Nutr 2011;52:408-413. 39. Nikaki K, Woodland P, Lee C, et al. Esophageal mucosal innerva-
23. Triggs JR, Kahrilas PJ. Editorial: symptom association probability during tion in functional heartburn: closer to healthy asymptomatic subjects
reflux testing-what is the gain? Aliment Pharmacol Ther 2018;47:1317- than to non-erosive reflux disease patients. Neurogastroenterol Motil
1338. 2019;31:e13667.
24. Ribolsi M, Guarino MPL, Tullio A, Cicala M. Post-reflux swallow- 40. Woodland P, Shen Ooi JL, Grassi F, et al. Superficial esophageal mucosal
induced peristaltic wave index and mean nocturnal baseline impedance afferent nerves may contribute to reflux hypersensitivity in nonerosive
predict PPI response in GERD patients with extra esophageal symp- reflux disease. Gastroenterology 2017;153:1230-1239.
toms. Dig Liver Dis 2020;52:173-177. 41. Nikaki K, Lee C, Ustaoglu A, et al. Esophageal mucosa innerva-
25. Frazzoni M, Manta R, Mirante VG, Conigliaro R, Frazzoni L, Melotti tion in children with nonerosive reflux disease. Am J Gastroenterol
G. Esophageal chemical clearance is impaired in gastro-oesophageal 2021;116:1727-1729.
reflux disease — a 24-h impedance-pH monitoring assessment. Neuro- 42. Zentilin P, Savarino V, Mastracci L, et al. Reassessment of the diagnostic
gastroenterol Motil 2013;25:399-406, e295. value of histology in patients with GERD, using multiple biopsy sites and
26. Martinucci I, de Bortoli N, Savarino E, et al. Esophageal baseline imped- an appropriate control group. Am J Gastroenterol 2005;100:2299-2306.
ance levels in patients with pathophysiological characteristics of functional 43. Savarino E, de Bortoli N, De Cassan C, et al. The natural history of
heartburn. Neurogastroenterol Motil 2014;26:546-555. gastro-esophageal reflux disease: a comprehensive review. Dis Esophagus
27. Frazzoni M, Savarino E, de Bortoli N, et al. Analyses of the post-reflux 2017;30:1-9.
swallow-induced peristaltic wave index and nocturnal baseline impedance 44. Yang M, Li ZS, Xu XR, et al. Characterization of cortical potentials
parameters increase the diagnostic yield of patients with reflux disease. evoked by oesophageal balloon distention and acid perfusion in patients
Clin Gastroenterol Hepatol 2016;14:40-46. with functional heartburn. Neurogastroenterol Motil 2006;18:292-299.
28. Mahoney LB, Rosen R. The spectrum of reflux phenotypes. Gastroen- 45. Kern MK, Birn RM, Jaradeh S, et al. Identification and characterization
terol Hepatol 2019;15:646-654. of cerebral cortical response to esophageal mucosal acid exposure and
29. Gilger MA, El-Serag HB, Gold BD, et al. Prevalence of endoscopic distention. Gastroenterology 1998;115:1353-1362.
findings of erosive esophagitis in children: a population-based study. J 46. Shapiro M, Green C, Bautista JM, et al. Functional heartburn patients
Pediatr Gastroenterol Nutr 2008;47:141-146. demonstrate traits of functional bowel disorder but lack a uniform increase
30. Ristic N, Milovanovic I, Radusinovic M, et al. The comparative analyses of chemoreceptor sensitivity to acid. Am J Gastroenterol 2006;101:1084-
of different diagnostic approaches in detection of gastroesophageal reflux 1091.
disease in children. PLoS One 2017;12:e0187081. 47. Savarino E, Pohl D, Zentilin P, et al. Functional heartburn has more in
31. Martinez SD, Malagon IB, Garewal HS, Cui H, Fass R. Non-erosive common with functional dyspepsia than with non-erosive reflux disease.
reflux disease (NERD)—acid reflux and symptom patterns. Aliment Gut 2009;58:1185-1191.
Pharmacol Ther 2003;17:537-545. 48. de Bortoli N, Frazzoni L, Savarino EV, et al. Functional heartburn over-
32. Nikaki K, Sifrim D. Pathophysiology of pediatric gastroesophageal re- laps with irritable bowel syndrome more often than GERD. Am J Gas-
flux disease: similarities and differences with adults. J Clin Gastroenterol troenterol 2016;111:1711-1717.
2022;56:99-113. 49. Fass R, Naliboff BD, Fass SS, et al. The effect of auditory stress on per-
33. Lawal A, Kern M, Sanjeevi A, et al. Neurocognitive processing of esoph- ception of intraesophageal acid in patients with gastroesophageal reflux
ageal central sensitization in the insula and cingulate gyrus. Am J Physiol disease. Gastroenterology 2008;134:696-705.
Gastrointest Liver Physiol 2008;294:G787-G794. 50. Naliboff BD, Mayer M, Fass R, et al. The effect of life stress on symp-
34. Herregods TV, Bredenoord AJ, Smout AJ. Pathophysiology of gastro- toms of heartburn. Psychosom Med 2004;66:426-434.
esophageal reflux disease: new understanding in a new era. Neurogastro- 51. Wright CE, Ebrecht M, Mitchell R, Anggiansah A, Weinman J. The
enterol Motil 2015;27:1202-1213. effect of psychological stress on symptom severity and perception in
35. Yang M, Li ZS, Chen DF, et al. Quantitative assessment and character- patients with gastro-oesophageal reflux. J Psychosom Res 2005;59:415-
ization of visceral hyperalgesia evoked by esophageal balloon distention 424.
and acid perfusion in patients with functional heartburn, nonerosive reflux 52. Yadlapati R, Tye M, Keefer L, Kahrilas PJ, Pandolfino JE. Psychosocial
disease, and erosive esophagitis. Clin J Pain 2010;26:326-331. distress and quality of life impairment are associated with symptom sever-
36. Patel A, Hasak S, Nix BD, Sayuk GS, Newberry RD, Gyawali CP. Ge- ity in PPI non-responders with normal impedance-pH profiles. Am J
netic risk factors for perception of symptoms in GERD: an observational Gastroenterol 2018;113:31-38.
cohort study. Aliment Pharmacol Ther 2018;47:289-297. 53. Kessing BF, Bredenoord AJ, Saleh CM, Smout AJ. Effects of anxiety

164 Journal of Neurogastroenterology and Motility

 Non-erosive Esophageal Phenotypes in Children

and depression in patients with gastroesophageal reflux disease. Clin Gas- heartburn. J Gastroenterol 2013;48:473-482
troenterol Hepatol 2015;13:1089-1095, e1. 58. Kessing BF, Bredenoord AJ, Weijenborg PW, Hemmink GJ, Loots
54. Bilgi MM, Vardar R, Yıldırım E, Veznedaroğlu B, Bor S. Prevalence CM, Smout AJPM. Esophageal acid exposure decreases intraluminal
of psychiatric comorbidity in symptomatic gastroesophageal reflux sub- baseline impedance levels. Am J Gastroenterol 2011;106:2093-2097.
groups. Dig Dis Sci 2017;62:984-993. 59. Ribolsi M, Emerenziani S, Borrelli O, et al. Impedance baseline and
55. Kandulski A, Jechorek D, Caro C, et al. Histomorphological differentia- reflux perception in responder and non-responder non-erosive reflux dis-
tion of non-erosive reflux disease and functional heartburn in patients ease patients. Scand J Gastroenterol 2012;47:1266-1273.
with PPI-refractory heartburn. Aliment Pharmacol Ther 2013;38:643- 60. Frazzoni L, Frazzoni M, de Bortoli N, et al. Postreflux swallow-induced
651. peristaltic wave index and nocturnal baseline impedance can link PPI-
56. Savarino E, Pohl D, Zentilin P, et al. Functional heartburn has more in responsive heartburn to reflux better than acid exposure time. Neurogas-
common with functional dyspepsia than with non-erosive reflux disease. troenterol Motil 2017;29:e13116.
Gut 2009;58:1185-1191. 61. Frazzoni M, de Bortoli N, Frazzoni L, Tolone S, Savarino V, Savarino E.
57. Savarino E, Zentilin P, Mastracci L, et al. Microscopic esophagitis distin- Impedance-pH monitoring for diagnosis of reflux disease: new perspec-
guishes patients with non-erosive reflux disease from those with functional tives. Dig Dis Sci 2017;62:1881-1889.

Vol. 29, No. 2 April, 2023 (156-165) 165