The brain–esophagus axis in subjects with and without obesity assessed by esophageal acid perfusion and functional brain imaging

Marcelo Ribeiroa,b, Cassiano Mateus Forcelinic, José Carlos Tomiozzo Jr.a,d, Ricardo Bernardi Sodere, Fernando Fornaria,c,f

Universidade Federal do Rio Grande do Sul, Porto Alegre-RS; Clínica KOZMA, Passo Fundo-RS; Universidade de Passo Fundo, Passo Fundo-RS; Atitus Educação, Passo Fundo-RS; Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre-RS, Brazil

aPrograma de Pós-Graduação: Ciências em Gastroenterologia e Hepatologia, Faculdade de Medicina, Universidade Federal do Rio Grande do Sul, Porto Alegre-RS (Marcelo Ribeiro, José Carlos Tomiozzo Jr., Fernando Fornari); bClínica KOZMA, Passo Fundo-RS (Marcelo Ribeiro); cFaculdade de Medicina, Universidade de Passo Fundo, Passo Fundo-RS (Cassiano Mateus Forcelini, Fernando Fornari); dFaculdade de Medicina, Atitus Educação, Passo Fundo-RS (José Carlos Tomiozzo Jr.); eInstituto do Cérebro, Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre-RS (Ricardo Bernardi Soder); fFaculdade de Odontologia, Programa de Pós-Graduação em Odontologia, Faculdade de Odontologia, Universidade de Passo Fundo, Passo Fundo-RS (Fernando Fornari), Brazil)

Correspondence to: Cassiano Mateus Forcelini, MD, PhD, Faculdade de Medicina (UPF), R. Teixeira Soares 817, Centro, CEP: 99010-080, Passo Fundo – RS, Brazil, e-mail: cmforcelini@gmail.com
Received 5 February 2023; accepted 24 May 2023; published online 3 July 2023
DOI: https://doi.org/10.20524/aog.2023.0818
© 2023 Hellenic Society of Gastroenterology

Abstract

Background Gastroesophageal reflux disease (GERD) has a complex pathophysiology and a heterogeneous symptom profile. The brain–esophageal axis in GERD has been studied with functional brain imaging during the last decades, but data from obese patients was just recently reported. A comparison of such a group with non-obese subjects is lacking in the literature. This study aimed to evaluate heartburn perception and brain connectivity responses during esophageal acid stimulation in subjects with and without obesity, controlling for the presence of typical reflux symptoms.

Methods In this cross-sectional study, 25 patients with obesity (body mass index ≥30 kg/m2) and 46 subjects without obesity underwent functional magnetic resonance imaging (fMRI) of the brain with esophageal water and acid perfusion. The fMRI paradigm and connectivity were assessed.

Results About two-thirds of the participants had reflux symptoms. Heartburn perception during fMRI did not differ between subjects with and without obesity. The presence of reflux symptoms was associated with lower activation in frontal brain regions during acid perfusion compared to water perfusion. Compared to subjects without obesity, patients with obesity presented significantly lower connectivity within the anterior salience network. Corrected clusters included left caudate, left putamen and left anterior cingulate gyrus.

Conclusions The brain–esophagus axis showed differences between subjects with and without obesity. Even without symptomatic differences following esophageal acid perfusion, patients with reflux symptoms showed less brain activation in frontal areas, while obese individuals presented lower connectivity within the anterior salience network.

Keywords Functional magnetic resonance imaging, brain activation, heartburn perception, gastroesophageal reflux disease

Ann Gastroenterol 2023; 36 (5): 504-510

Introduction

Gastroesophageal reflux disease (GERD) is a relevant condition with a high prevalence, complex pathophysiology and a heterogeneous symptom profile [1]. Its modern diagnosis relies on the application of consensual parameters on esophageal testing for conclusive identification of GERD phenotypes [2]. The prevalence of GERD is higher in the population with obesity [3]. This results from conditions related to obesity, including eating behaviors, higher gastric pressure, visceral fat-related inflammation [4] and proximal dislocation of the gastric pocket [5]. In addition, hiatal hernia and transient lower esophageal sphincter relaxations are more frequent in obese patients [6,7].

Esophageal sensitivity to reflux of gastric contents is based on the activation of chemo- and mechanoreceptors [8], with variations according to GERD phenotypes. Overall, patients with functional heartburn show higher sensitivity to chemical and mechanical stimuli than other GERD phenotypes [9]. In obese individuals, esophageal mechanisms such as epithelial resistance and visceral sensitivity are less understood [10]. Although most studies have described esophageal hyposensitivity with high preoperative rates of reflux esophagitis in oligosymptomatic candidates for bariatric surgery [11,12], there is some evidence suggesting a visceral hypersensitivity associated with obesity [13].

The application of validated questionnaires for GERD identification is widely accepted in clinical research, despite their acknowledged flaws [14]. Such instruments can translate the subjectivity of GERD manifestations into objective symptom scores, guiding both diagnosis and responses to therapeutical strategies. However, GERD questionnaires have limited accuracy when compared to objective reflux testing [2], with sensitivity and specificity inferior to 70%.

New technologies have enabled the evaluation of brain responses related to visceral stimulation [15]. Functional magnetic resonance imaging (fMRI) has been used to assess the brain–esophagus axis and the relationship with perception of heartburn in volunteers and patients with GERD [16,17]. This tool allows both task analysis as a response to esophageal acid stimulation and the assessment of task-residual connectivity [18]. Studies with this technology, combined with esophageal acid perfusion, have described different brain activation patterns among the GERD subtypes [19].

The evaluation of the brain–esophageal axis was recently characterized in patients with obesity [20,21]. We aimed to assess heartburn perception and brain connectivity responses during esophageal acid stimulation in subjects with and without obesity, controlling for the presence of reflux symptoms.

Patients and methods

Participants

In this cross-sectional study, which included a retrospective analysis of a prospectively built database, 26 patients with obesity (body mass index [BMI] ≥30 kg/m2) and 46 subjects without obesity were recruited and studied from 2010-2015, following a unique protocol for data acquisition [22]. Among the obese subjects there were 9 healthy volunteers, 15 patients with troublesome heartburn and normal esophageal mucosa on endoscopy, and 2 patients with troublesome heartburn and erosive esophagitis. The non-obese subjects included 17 healthy volunteers, 18 patients with troublesome heartburn and normal esophageal mucosa on endoscopy, and 11 patients with troublesome heartburn and erosive esophagitis.

The inclusion criteria for all participants were: (1) older than 18 years; (2) right-handed; and (3) consent to participate. Exclusion criteria were: (1) uncompensated clinical morbidities; (2) gastroesophageal surgery; (3) claustrophobia; (4) regular use of acid-suppressive medications or drugs acting on the central nervous system; (5) pregnancy; and (6) use of metal orthodontic appliances. Reflux symptoms were characterized in the presence of troublesome heartburn and/or acid regurgitation evaluated with a validated GERD symptoms questionnaire [23]. All participants apart from the healthy volunteers without obesity underwent upper digestive endoscopy prospectively or in the last 6 months from date eligibility. Reflux esophagitis was identified according to Los Angeles classification (grades B, C and D). Anxiety and depression scores were obtained with the Hamilton Anxiety Scale and Beck Depression Inventory version 1A, respectively [24,25]. All participants gave their informed consent and the study was approved (numbers 150/2009 and 383/2011) by the local Ethics Committee.

Study protocol

Esophageal acid stimulation

A mono-lumen catheter was inserted trans-nasally and positioned at 5 cm proximally to the lower esophageal sphincter according to a previously mapped pH turning point, as described elsewhere [26]. After approximately 15 min from catheter placement, the patients were placed in the fMRI machine. Subjects were instructed to lie still during the exam and signal with one finger of the left hand when heartburn was perceived. The experimental protocol included an anatomical scan followed by 2 periods of 5 min each, with esophageal perfusion at 1 mL/min. The first perfusion was performed with distilled water (pH 6.5) and the second with a solution based on hydrochloric acid (pH 1.5). The acid solution was prepared by a compounding pharmacy, with hydrochloric acid added to distilled water until its pH stabilized at 1.5. In both perfusion periods, we acquired functional images of the brain. Although the perfusion order was not randomized, the patients were blinded to the solution type sequence.

fMRI data acquisition

Magnetic resonance images were acquired at Clínica Kozma (Passo Fundo, RS, Brazil) using a 1.5 Tesla Magnetom Avanto (Siemens AG, Munich, Germany), equipped with a standard 8-channel head coil. High-resolution tridimensional anatomic images were obtained from 176 sagittal spoiled gradient recalled sequence slices with a voxel size of 1.0 × 1.0 × 1.0 mm, for further co-registration with functional images. For functional acquisition, echo planar images were obtained in a mosaic composed by 36 axial slices of 64 × 64 pixels over a 220 mm field of view, with a voxel size of 3.0 × 3.0 × 3.75 mm, a repetition time of 4270 msec and an echo time of 50 msec.

MRI preprocessing and first-level analysis

MRI functional images were preprocessed using the FMRIB Software Library (FSL) [27]. Brain extraction was performed using the Brain Extraction Tool [28]. Functional data preprocessing was then carried out using the FEAT program standard pipeline for first-level analysis [29]. Data underwent motion correction, slice timing correction and Gaussian kernel smoothing full width at half maximum of 4.0 mm. Functional data were co-registered with a subject-respective T1 anatomical image. Anatomical data were registered in standard MNI152 space, using FLIRT with 12 degrees of freedom, and the resulting transformation matrix was applied to warp co-registered functional data into standardized space. We then, modeled a task time series within FEAT [30], modeling the hemodynamic response function assuming a sinusoidal function for water and acid perfusion effects on signal variance.

Task-residual functional connectivity (trFC) analysis

After first-level analysis, the resulting residuals (effects not explained by task), underwent independent component analysis (ICA) carried out with MELODIC [18]. ICA analysis decomposed each participant’s functional data into spatial and temporal components, and concatenated all subjects’ components using multi-session temporal concatenation to generate group-spatial maps. Non noise-related components of interest were selected according to a functional network atlas [31]. From the group-selected components we created masks using a threshold of z>2.3 that would be utilized further in group comparisons.

Statistical analysis

Group comparisons between subjects for task data were made using the FLAME1 part of FEAT for higher-level analysis [32], which uses mixed effects modeling to estimate signal parameters. We account for normalized BMI as a continuous variable and the presence of reflux symptoms as a dummy variable in our task model; cluster correction for multiple comparisons was than performed, using a threshold of z>2.3. trFC group comparisons were carried out in FSL using dual regression and the program Randomise. First, group-spatial-maps were regressed for each specific subject time series, after those time series had been regressed into the same 4D data to generate subject-specific spatial maps. Eventually, Randomise was used to compare groups. We accounted for normalized BMI as a continuous variable and reflux symptoms status as a dummy variable in the first model, and in the second model, we accounted for obesity (BMI >30 kg/m2) as a dummy variable (2-sample independent t-test). Randomise was carried out using 5000 permutations, and cluster correction for multiple comparisons; a P-value <0.05 for corrected clusters was considered significant. From significant clusters, we created region of interest (ROI) masks and warped them to subject-specific native space, to extract connectivity z-score values from those ROIs. Demographic and clinical characteristics were compared between obese and non-obese using the Student’s t- and χ-square tests for quantitative and qualitative variables, respectively. In the case of asymmetric quantitative data according to the Kolmogorov-Smirnov test, the Mann-Whitney U test was used. Analyses were performed with GraphPad Prism version 5 (GraphPad Software Inc, San Diego CA, USA) and Statistical Package for Social Sciences version 16.0 (SPSS Inc, Chicago, IL, USA). The P-value defined as significant was 0.05.

Results

Participants

A total of 72 subjects were enrolled in the study. One participant was excluded because of an artifact on the fMRI data. The 71 remaining participants included 25 obese patients and 46 without obesity (Table 1). Women and men were equally distributed in both groups, whereas participants with obesity were slightly older. The morbidity variables, including reflux symptoms, depression and anxiety, were all similar between groups. Troublesome reflux symptoms and/or reflux esophagitis were found in approximately 65% of the participants in both groups.

Table 1 Characteristics of the participants (n = 71)

thumblarge

Heartburn perception during fMRI

The perception of heartburn did not differ between subjects with and without obesity, being signaled by approximately 40% of the participants (Table 2). A subgroup analysis of participants with reflux symptoms (n=46) also did not show differences in heartburn perception between groups. Among 6 participants with obesity in which the heartburn signaling was discriminated according to perfusion type, 24% of heartburn episodes were signaled during water perfusion and 76% during acid perfusion in the esophagus.

Table 2 Heartburn perception (and score) during fMRI with esophageal perfusion (5 min water and 5 min acid) in subjects with and without obesity (all participants, regardless of reflux symptoms/RE), and in obese and non-obese patients with reflux symptoms/RE (Los Angeles B, C or D)

thumblarge

Brain responses to esophageal acid stimulation

Task activation

The presence of reflux symptoms was associated with lower brain activation in frontal brain areas during acid perfusion compared to water perfusion, after controlling for BMI (Table 3). Significant clusters included left and right frontal orbital cortex, right inferior frontal gyrus and left frontal pole. BMI was not associated with significant changes in brain activation, after adjustment for reflux symptom status.

Table 3 Significant activation clusters, associated with the presence of typical reflux symptoms (control > presence of symptoms), comparing acid perfusion with water perfusion

thumblarge

trFC

In the model using BMI adjusted for reflux symptom status, no significant changes were found in any functional network. In the model with dichotomous BMI (obese vs. nonobese), patients with obesity presented significantly lower connectivity within the anterior salience network (Fig. 1, 2), in the left caudate, left putamen and left anterior cingulate gyrus regions. Cluster information with mean normalized connectivity for each group is shown in Table 4. There were no significant differences between groups within the default mode network, sensorimotor network, executive network or primary visual network.

thumblarge

Figure 1 Differences in task-residual functional connectivity between subjects with and without obesity within the anterior salience network. (A) Left anterior cingulate gyrus cluster. (B) Putamen and caudate cluster. Blue codifies lower values in patients with obesity

thumblarge

Figure 2 Mean connectivity values within the anterior salience network for each group in each cluster. Left image shows mean connectivity in the putamen and caudate nuclei. Right image shows mean connectivity for the left anterior cingulate gyrus (LACG). Patients with obesity correspond to the blue boxplot and controls to the orange

Table 4 Significant trFC clusters, associated with obesity status within the anterior salience network (non-obese>obese)

thumblarge

Discussion

In this study involving participants with and without obesity, with a variety of GERD phenotypes, we investigated whether heartburn perception and brain responses during esophageal acid stimulation were different in relation to obesity. Both task analysis, as a response to esophageal acid stimulation, and task-residual connectivity were obtained with fMRI. As far as we know, this is the first study addressing the brain–esophagus axis in subjects with and without obesity in the context of reflux symptoms.

The main findings were: (a) overall perception of heartburn during esophageal acid perfusion was similar between groups; (b) among patients with reflux symptoms, the perception of heartburn also did not differ, regardless of obesity; (c) compared to controls, patients with obesity presented significantly lower connectivity within the anterior salience network, including the left caudate, left putamen and left anterior cingulate gyrus; (d) patients with reflux symptoms presented lower brain activation related to esophageal acid perfusion in frontal brain regions; and (e) participants with and without obesity were balanced in terms of depression and anxiety scores, softening the influence of these emotional conditions on results.

Visceral hypersensitivity is more commonly reported in patients with functional gastrointestinal diseases, such as functional heartburn [33]. This mechanism, represented by the brain–esophagus axis, contributes to functional esophageal pain and consists of a complex interaction between peripheral nerves, superior cortical centers, a descending inhibitory pathway and psychological modulation triggered by the stimulus [34]. Few studies have addressed the role of obesity in esophageal sensitivity in the context of GERD. Mercer et al compared asymptomatic subjects with obesity and lean volunteers and found esophageal hypersensitivity to acid perfusion in the group with obesity [13]. In contrast, Ortiz et al reported that asymptomatic GERD is more common than the symptomatic pattern in patients with severe obesity (BMI ≥40 kg/m2) [11].

In the present study, the perception of heartburn during esophageal acid stimulation was similar in both groups. Acid perfusion in the distal esophagus lasted 5 min using a solution with very low pH (~ 1.5) and was able to trigger heartburn in approximately 40% of the participants. This model of esophageal stimulation followed the patterns proposed by Kern et al, pioneers in the evaluation of the brain–esophageal axis in patients with GERD [16]. The use of a multimodal esophageal stimulation paradigm [35], including thermal, mechanical, electrical and chemical (acidic solution) stimulations, could provide additional data regarding esophageal sensitivity.

The limbic system, which includes the insular, cingulate, amygdala, thalamus, hypothalamus and prefrontal cortex, was reported to participate in the processing of afferent sensory inputs of the esophagus [16,36]. In particular, the insula and cingulate play a role in the integration of autonomic, affective, cognitive and motor responses to sensory signals originating from the gastrointestinal tract [37]. The prefrontal lobe cortex is the senior center of algesthesia. Somatic and visceral sensory information converge on the insula, which can manage sensation and emotion. In fact, studies comparing GERD patients with controls have demonstrated abnormal default mode network and insular cortex functional connectivity in the patients, reinforcing the role of such areas in the esophageal perception of reflux [38,39]. However, none of these studies specifically examined the influence of obesity on the results.

In our study, patients with obesity presented significantly lower connectivity within the anterior salience network, including the left caudate, left putamen and left anterior cingulate gyrus, compared to subjects without obesity. This finding may be an initial explanation for asymptomatic GERD being more common than the symptomatic pattern in those with severe obesity [11]. On the other hand, lower connectivity in obese subjects was already reported during the resting state or milkshake consumption, suggesting that this finding may be a non-specific response due to obesity [40,41]. The anterior salience network is part of the “default mode of brain function”, defined as a task-negative network, the activity of which is greater at rest than under various goal-directed tasks [42]. In addition, participants with reflux symptoms presented lower brain activation related to acid perfusion in frontal brain regions compared to participants without reflux symptoms. However, in the model using BMI adjusted for reflux symptom status, no significant change was found in functional networks.

Most neuroimaging studies of the brain–esophageal axis are based on intermittent or continuous esophageal stimulation. Though these are not stimulus-specific, anticipation of stimuli can trigger similar activity, and repeated activations can result in adaptation [43]. Through the application of a single acidic perfusion, we were able to minimize these effects. Moreover, the resting state networks identified through ICA point to intriguing group differences in the central response to esophageal stimulation in subjects with and without obesity. Further studies including both groups are still needed to understand the complex interactions of the brain–esophagus axis [15].

We acknowledge that the study had some limitations. The task analysis was not properly designed to generate fMRI output, since each esophageal stimulus was presented only once to participants, in a single fMRI session, as well as for a long time (300 sec). Regarding trFC analysis, we did not account for age and sex in our model; since these are important confounders in neuroimaging studies and our study sample was heterogeneous, this may have influenced the results. There was a significant age difference between the groups. However, the age difference was just 5.8 years, less than in studies that reported age-dependent changes in functional connectivity within the anterior salience network [44,45]. The periods of perfusion were short (only 5 min), and the severity of heartburn was not quantified, limiting the assessment of esophageal sensitivity. Nevertheless, the inclusion of different GERD phenotypes in young adult participants with and without obesity translates into a common type of patient in clinical practice, increasing the generalizability of our results.

In summary, in the presence of reflux symptoms, esophageal sensitivity evoked by acid perfusion was equal in subjects with and without obesity in terms of heartburn perception, but accompanied by specific changes in brain responses, especially in those with obesity. Further research into this matter is warranted to explore possible clinical applications of the findings.

Summary Box

What is already known:

  • The prevalence of gastroesophageal reflux disease (GERD) is higher in the population with obesity

  • Functional magnetic resonance imaging (fMRI) has been used to assess the brain–esophagus axis and the relationship with the perception of heartburn in volunteers and patients with GERD

  • Evaluation of the brain–esophageal axis is lacking among patients with obesity

What the new findings are:

  • In the presence of reflux symptoms, brain activation is lower in frontal areas

  • Heartburn perception during fMRI does not differ between subjects with and without obesity

  • Obesity is related to lower connectivity within the anterior salience network

References

1. Xiao YL, Zhou LY, Hou XH, Li YQ, Zou DW, Chen MH;Chinese Society of Gastroenterology. Chinese expert consensus on gastroesophageal reflux disease in 2020. J Dig Dis 2021;22:376-389.

2. Gyawali CP, Kahrilas PJ, Savarino E, et al. Modern diagnosis of GERD:the Lyon Consensus. Gut 2018;67:1351-1362.

3. El-Serag HB, Graham DY, Satia JA, Rabeneck L. Obesity is an independent risk factor for GERD symptoms and erosive esophagitis. Am J Gastroenterol 2005;100:1243-1250.

4. Tilg H, Moschen AR. Visceral adipose tissue attacks beyond the liver:esophagogastric junction as a new target. Gastroenterology 2010;139:1823-1826.

5. Robertson EV, Derakhshan MH, Wirz AA, et al. Central obesity in asymptomatic volunteers is associated with increased intrasphincteric acid reflux and lengthening of the cardiac mucosa. Gastroenterology 2013;145:730-739.

6. Wilson LJ, Ma W, Hirschowitz BI. Association of obesity with hiatal hernia and esophagitis. Am J Gastroenterol 1999;94:2840-2844.

7. Schneider JH, Küper M, Königsrainer A, Brücher B. Transient lower esophageal sphincter relaxation in morbid obesity. Obes Surg 2009;19:595-600.

8. Fass R, Naliboff B, Higa L, et al. Differential effect of long-term esophageal acid exposure on mechanosensitivity and chemosensitivity in humans. Gastroenterology 1998;115:1363-1373.

9. Yang M, Li ZS, Chen DF, et al. Quantitative assessment and characterization of visceral hyperalgesia evoked by esophageal balloon distention and acid perfusion in patients with functional heartburn, nonerosive reflux disease, and erosive esophagitis. Clin J Pain 2010;26:326-331.

10. Tack J, Pandolfino JE. Pathophysiology of gastroesophageal reflux disease. Gastroenterology 2018;154:277-288.

11. Ortiz V, Ponce M, Fernández A, et al. Value of heartburn for diagnosing gastroesophageal reflux disease in severely obese patients. Obesity (Silver Spring) 2006;14:696-700.

12. Sharara AI, Rustom LBO, Bou Daher H, et al. Prevalence of gastroesophageal reflux and risk factors for erosive esophagitis in obese patients considered for bariatric surgery. Dig Liver Dis 2019;51:1375-1379.

13. Mercer CD, Wren SF, DaCosta LR, Beck IT. Lower esophageal sphincter pressure and gastroesophageal pressure gradients in excessively obese patients. J Med 1987;18:135-146.

14. Bolier EA, Kessing BF, Smout AJ, Bredenoord AJ. Systematic review:questionnaires for assessment of gastroesophageal reflux disease. Dis Esophagus 2015;28:105-120.

15. Mayer EA. The role of gut-brain interactions in influencing symptoms of irritable bowel syndrome. Gastroenterol Hepatol (N Y) 2018;14:44-46.

16. Kern M, Hofmann C, Hyde J, Shaker R. Characterization of the cerebral cortical representation of heartburn in GERD patients. Am J Physiol Gastrointest Liver Physiol 2004;286:G174-G181.

17. Lawal A, Kern M, Sanjeevi A, et al. Neurocognitive processing of esophageal central sensitization in the insula and cingulate gyrus. Am J Physiol Gastrointest Liver Physiol 2008;294:G787-G794.

18. Smith SM, Jenkinson M, Woolrich MW, et al. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage 2004;23 Suppl 1:S208-S219.

19. Wang K, Duan LP, Zeng XZ, Liu JY, Xu-Chu W. Differences in cerebral response to esophageal acid stimuli and psychological anticipation in GERD subtypes - An fMRI study. BMC Gastroenterol 2011;11:28.

20. Mayer EA, Labus J, Aziz Q, et al. Role of brain imaging in disorders of brain-gut interaction:a Rome Working Team Report. Gut 2019;68:1701-1715.

21. Ribeiro M, Forcelini CM, Navarini D, Soder RB, Fornari F. Disruption of the brain-esophagus axis in obese patients with heartburn. Dis Esophagus 2022;35:doac021.

22. Forcelini CM, Tomiozzo JC Jr, FarréR, et al. Effect of nortriptyline on brain responses to painful esophageal acid infusion in patients with non-erosive reflux disease. Neurogastroenterol Motil 2014;26:187-195.

23. Fornari F, Gruber AC, Lopes AB, Cecchetti D, de Barros SG. [Symptom's questionnaire for gastroesophageal reflux disease]. Arq Gastroenterol 2004;41:263-267.

24. Hamilton M. The assessment of anxiety states by rating. Br J Med Psychol 1959;32:50-55.

25. Beck AT, Epstein N, Brown G, Steer RA. An inventory for measuring clinical anxiety:psychometric properties. J Consult Clin Psychol 1988;56:893-897.

26. Liell TP, Tomiozzo JC Jr, Denti F, de Lima LA, Fornari F. Determination of pH turning point with pH mapping of the gastroesophageal junction:an alternative technique to orientate esophageal pH monitoring. Dis Esophagus 2011;24:305-311.

27. Jenkinson M, Beckmann CF, Behrens TE, Woolrich MW, Smith SM. FSL. Neuroimage 2012;62:782-790.

28. Smith SM. Fast robust automated brain extraction. Hum Brain Mapp 2002;17:143-155.

29. Woolrich MW, Ripley BD, Brady M, Smith SM. Temporal autocorrelation in univariate linear modeling of FMRI data. Neuroimage 2001;14:1370-1386.

30. Woolrich MW, Behrens TE, Smith SM. Constrained linear basis sets for HRF modelling using Variational Bayes. Neuroimage 2004;21:1748-1761.

31. Altmann A, Ng B, Landau SM, Jagust WJ, Greicius MD;Alzheimer's Disease Neuroimaging Initiative. Regional brain hypometabolism is unrelated to regional amyloid plaque burden. Brain 2015;138:3734-3746.

32. Woolrich MW, Behrens TE, Beckmann CF, Jenkinson M, Smith SM. Multilevel linear modelling for FMRI group analysis using Bayesian inference. Neuroimage 2004;21:1732-1747.

33. Aziz Q, Fass R, Gyawali CP, Miwa H, Pandolfino JE, Zerbib F. Functional Esophageal Disorders. Gastroenterology 2016;150:1368-1379.

34. Fass R, Tougas G. Functional heartburn:the stimulus, the pain, and the brain. Gut 2002;51:885-892.

35. Broers C, Boecxstaens V, Deloose E, Tack J, Pauwels A. The optimal order of stimulation modalities and reproducibility of the multimodal esophageal stimulation paradigm. Neurogastroenterol Motil 2019;31:e13475.

36. Hobson AR, Aziz Q. Brain processing of esophageal sensation in health and disease. Gastroenterol Clin North Am 2004;33:69-91.

37. Derbyshire SW. A systematic review of neuroimaging data during visceral stimulation. Am J Gastroenterol 2003;98:12-20.

38. Sun H, Chen Y, Zhao X, et al. Abnormal activity of default mode network in GERD patients. BMC Neurosci 2013;14:69.

39. Siwiec RM, Babaei A, Kern M, Samuel EA, Li SJ, Shaker R. Esophageal acid stimulation alters insular cortex functional connectivity in gastroesophageal reflux disease. Neurogastroenterol Motil 2015;27:201-211.

40. Geha P, Cecchi G, Todd Constable R, Abdallah C, Small DM. Reorganization of brain connectivity in obesity. Hum Brain Mapp 2017;38:1403-1420.

41. Doornweerd S, van Duinkerken E, de Geus EJ, Arbab-Zadeh P, Veltman DJ, IJzerman RG. Overweight is associated with lower resting state functional connectivity in females after eliminating genetic effects:A twin study. Hum Brain Mapp 2017;38:5069-5081.

42. Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A 2005;102:9673-9678.

43. Frøkjær JB, Olesen SS, Graversen C, Andresen T, Lelic D, Drewes AM. Neuroimaging of the human visceral pain system—a methodological review. Scand J Pain 2018;2:95-104.

44. Goldstone A, Mayhew SD, Przezdzik I, Wilson RS, Hale JR, Bagshaw AP. Gender specific re-organization of resting-state networks in older age. Front Aging Neurosci 2016;8:285.

45. Long Z, Cheng F. Age effect on functional connectivity changes of right anterior insula after partial sleep deprivation. Neuroreport 2019;30:1246-1250.

Notes

Conflict of Interest: None

Funding: This study was partially supported by Clínica Kozma and Endopasso