Gyawali CP, Marchetti L, Rogers BD, et al. The Lyon Score: A Novel Reflux Scoring System Based on the Lyon Consensus 2.0 That Associates With Treatment Outcome From Antireflux Therapy. The American Journal of Gastroenterology 120(5):p 1009-18, May 2025

Авторы: Gyawali C.P. / Marchetti L.  / Rogers B.D. / Chan W.W. / Wong M.-W. / Visaggi P. / Rengarajan A. / Carlson D.A. / Savarino E.V. / de Bortoli N. / Chen Ch.-L.  / Pandolfino J.E.

The Lyon Score: A Novel Reflux Scoring System Based on the Lyon Consensus 2.0 That Associates With Treatment Outcome From Antireflux Therapy

C. Prakash Gyawali, MD, FRCP, FACG¹, Lorenzo Marchetti, MD^1,2, Benjamin D. Rogers, MD, MS^1,3, Walter W. Chan, MD, MPH, FACG⁴, Ming-Wun Wong, MD⁵, Pierfrancesco Visaggi, MD⁶, Arvind Rengarajan, MD¹, Dustin A. Carlson, MD⁷, Edoardo Savarino, MD, PhD⁸, Nicola de Bortoli, MD⁶, Chien-Lin Chen, MD⁵ and John Pandolfino, MD, FACG<sup>7


1</sup>Division of Gastroenterology, Washington University School of Medicine, St. Louis, Missouri, USA; ²Department of Digestive Diseases, Campus Bio Medico University of Rome, Rome, Italy; ³Division of Gastroenterology, Hepatology, and Nutrition, University of Louisville School of Medicine, Louisville, Kentucky, USA; ⁴Division of Gastroenterology, Hepatology and Endoscopy, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, USA; ⁵Department of Medicine, Division of Gastroenterology and Hepatology, Hualien Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation and Tzu Chi University, Hualien, Taiwan; ⁶Department of Translational Research and New Technologies in Medicine and Surgery, Division of Gastroenterology, University of Pisa, Pisa, Italy; ⁷Kenneth C. Griffin Esophageal Center, Northwestern Medicine, Division of Gastroenterology and Hepatology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, USA; ⁸Gastroenterology Unit, Department of Surgery, Oncology and Gastroenterology, University of Padua, Padua, Italy. Correspondence: C. Prakash Gyawali, MD, FRCP, FACG. E-mail: cprakash@wustl.edu.

Abstract

INTRODUCTION: We explored if a score derived from parameters from esophageal testing could increase confidence in diagnosing conclusive gastroesophageal reflux disease and in predicting outcome.

METHODS: A prediction score was developed using metrics based on Lyon Consensus 2.0 thresholds extracted from endoscopy and pH-impedance monitoring. The Lyon score was the sum of weighted scores derived from a logistic regression model. The outcome was response to antireflux therapy, defined as 50% reduction in global symptoms on validated questionnaires. An existing database of endoscopy-negative patients with typical reflux symptoms undergoing esophageal testing from 2 centers (Europe and the United States) constituted the developmental cohort, while 2 separate cohorts (Europe and Asia) served as validation cohorts. Receiver operating characteristics analysis determined performance of the Lyon score in predicting treatment response.

RESULTS: In 281 developmental cohort patients (median age 53 years, 57.7% female), the Lyon score demonstrated an area under the curve (AUC) of 0.819 in predicting 50% symptom improvement (P < 0.001) on receiver operating characteristics, with an optimal threshold of 6.25 (sensitivity 81.2%, specificity 73.4%). Of the individual components, only acid exposure time (AUC 0.799, P < 0.001), mean nocturnal baseline impedance (AUC 0.785, P < 0.001), and reflux episodes (AUC 0.764, P < 0.001) approached the Lyon score performance. The Lyon score segregated treatment response in both the European (AUC 0.908, P < 0.001) and Asian validation cohorts (AUC 0.637, P < 0.001) and outperformed the DeMeester score in sensitivity for predicting outcome in the developmental and Asian validation cohorts.

DISCUSSION: The novel Lyon score segregates reflux phenotypes and identifies likelihood of symptom response from antireflux therapy. 

Keywords: pH-impedance monitoring; upper endoscopy; acid exposure time


INTRODUCTION

When patients with esophageal symptoms remain symptomatic despite empiric trials of proton-pump inhibitors (PPIs), esophageal testing can provide objective evidence for or against ongoing gastroesophageal reflux disease (GERD) (1,2). In patients with no prior GERD evidence, endoscopy and ambulatory reflux monitoring are performed off therapy to determine if GERD exists (3). The updated Lyon Consensus 2.0 describes testing methodology based on presentation and provides designation of test findings as conclusive, inconclusive, supportive, or not supportive of a GERD diagnosis (4).

Conclusive evidence for GERD consists of advanced esophagitis grade on endoscopy (Los Angeles grade B or higher) and abnormal reflux burden on ambulatory reflux monitoring (4). Esophageal reflux burden can be quantified using acid exposure time (AET), numbers of reflux episodes (RE), and mean nocturnal baseline impedance (MNBI), each of which has defined thresholds for physiologic, borderline, and pathologic reflux evidence (4). Alternate parameters such as reflux-symptom association (RSA) on ambulatory reflux monitoring and hiatus hernia on endoscopy and high-resolution manometry (HRM) provide supportive evidence for reflux. Each of these parameters has varying importance in establishing conclusive reflux evidence. While a scoring system based on acid-based parameters (DeMeester score) performs well in identifying patients who need antireflux management and predicts outcome (5), combinations of abnormal metrics from pH-impedance monitoring and endoscopy have potential to provide even higher confidence in segregating conclusive reflux from functional esophageal syndromes, and lead to better symptom reduction following medical or surgical antireflux therapy.

We hypothesized that a scoring system created to stratify degrees of confidence in results of each diagnostic test could associate with GERD phenotypes and symptom response from GERD management. Therefore, in this proof of concept study, our aim was to determine if a novel reflux score (the Lyon Score) designed to quantify reflux burden using Lyon Consensus 2.0 thresholds can phenotype reflux syndromes and predict symptom response from antireflux therapy by comparing outcomes to the DeMeester score calculated from the same pH-impedance studies.

METHODS

An existing database of patients with typical reflux symptoms (heartburn, regurgitation, esophageal chest pain) evaluated with manometry and pH impedance monitoring off therapy for persisting reflux symptoms despite normal endoscopy from 2 centers (one each in Europe and the United States) was used for the initial development of the Lyon score. Eligible patients needed to have typical reflux symptoms, and have undergone esophageal testing in the setting of incomplete symptom response from antireflux medications, or patients in whom reflux metrics needed to be quantified to determine need for ongoing GERD management, with follow-up available for assessment of symptom response to GERD therapy. Included patientswere required to have completed HRM (Medtronic, Duluth, GA) and ambulatory pH-impedance monitoring (Diversatek, Boulder, CO) off antisecretory therapy after documentation of normal esophageal mucosa (without erosive esophagitis) on endoscopy. This was a convenience cohort, as symptoms and esophageal test findings had already been collected as part of a previous report, with response to antireflux therapy having been assessed after completion of all diagnostic testing, and data collected for prediction analysis being entirely agnostic of the outcome at the time of collection (6). Patients with diagnoses of achalasia spectrum disorders, eosinophilic esophagitis (biopsies performed at discretion of original investigators), prior foregut surgery, inadequate studies (equipment malfunction, poor study quality), and incomplete data were excluded. The prediction score was validated in 2 cohorts of consecutive patients with typical reflux symptoms from a European center (University of Pisa, Pisa, Italy) and an Asian center (Buddhist Tzu Chi Hospital, Hualien, Taiwan), all of whom had testing that included endoscopy, manometry and pH-impedance monitoring off therapy, and similar inclusions and exclusions, except that patients with esophagitis were not excluded. These 2 centers were chosen to validate the Lyon score in 2 unique patient care pathways that differed fromthe developmental cohort: testing for persisting heartburn despite empiric therapy in the European center and up-front testing before therapy in a regurgitation predominant population at the Asian center. Patients or the public were not involved in the design of this retrospective report. The original data collection projects were approved by institutional review boards at Washington University in St. Louis, University of Padua, University of Pisa, and Buddhist Tzu Chi Hospital. Since this study involved post hoc analysis of preexisting deidentified data sets with no links to original patients or protected health information, additional institutional review board approval was not deemed necessary.

Lyon score
The Lyon score was derived by ascribing weighted numerical values using a logistic regression model to the Lyon Consensus 2.0 diagnostic criteria for GERD on endoscopy and reflux monitoring (4); the sum of these weighted numerical values formed the Lyon score (Table 1), with a range of 0–18.5. The review of studies evaluating LA grades of esophagitis (7), AET (8,9), REs (10), MNBI (11), RSA (12,13) and hiatus hernia (14) in relationship to GERD treatment outcomes indicate that these parameters are critical to modern GERD diagnosis as described in the Lyon Consensus 2.0 (4). Metrics from the developmental cohort were used for creating the Lyon score; scores for LA grades of esophagitis on endoscopy were compared with known acid exposure data to generate weighted esophagitis scores (15). An investigator not involved in creation of the original data sets (L.M.) used this scoring system to assign and compute scores for each patient in each cohort from metrics available in the data sets. 

Endoscopy
All patients underwent endoscopy off-medication before esophageal physiologic testing in both the developmental and validation cohorts. The developmental cohort only included patients with no evidence of esophagitis on endoscopy; hence, endoscopy scores were consistently 0 in this cohort. The validation cohorts included patients with esophagitis, characterized according to the LA grading system. The presence and size of hiatus hernia was extracted from endoscopy and corroborated with HRM reports when available.

Ambulatory reflux monitoring
pH-impedance monitoring was performed using HRM localization of the LES to position the pH sensor 5 cm proximal to the LES. All studies were performed off antisecretory therapy, and patients stopped PPI therapy for 7 days, anti-histamine-2 receptor antagonists for 3 days, and antacids for 24 hours before reflux monitoring. Analysis of pH-impedance studies was performed using commercially available software (Sandhill Scientific, now Diversatek, Boulder, Colorado). After excluding meal times, total distal AET was extracted and recorded from the pH-impedance studies. After assessing the mean value of impedance of three 10-minute time periods during the night with no swallows or REs in the distal most distal channel, MNBI was manually calculated as the mean of these 3 measurements (16). REs were automatically identified by the software and then manually reviewed at each participating center using Wingate Consensus criteria by investigators with prior experience and expertise in analysis of more than 400 pH impedance studies each (17). A positive RSA was defined by either symptom index >50% or symptom-association probability >95% (3). DeMeester scores were calculated using software embedded in automated software analysis.

Symptom response
All patients completed symptom questionnaires before HRM and rated their overall esophageal symptoms over the previous 2 weeks on a 10 cmvisual analog scale (global symptom severity), as previously reported (9). The same questionnaires were readministered to eligible patients on follow-up. Symptom response was assessed using the change in global symptom scores from visual analog scales extracted from patient report questionnaires over follow-up; a 50% reduction in global symptoms constituted a response to antireflux therapy, both medical management and antireflux surgery (ARS).

High-resolution manometry
HRM studies were performed using the HRM system available at each study center. HRM catheters with high-fidelity circumferential sensors 1 cm apart was passed through the anesthetized nasal channel (using viscous lidocaine) by an experienced operator and positioned with 3 distal sensors in the stomach whenever possible, using previously described methods (18). All HRM studies were performed in a supine, semi-recumbent position, using test swallows of 5 mL of room temperature water every 20–30 seconds until 10 swallows were performed.

For the purpose of this study, HRM studies were only used to corroborate esophagogastric junction (EGJ) morphology, i.e., presence and size of hiatus hernia. Esophageal body characteristics from HRM were not included as part of the Lyon score.

Statistical analysis
Data are reported as median (interquartile range) unless otherwise indicated. Categorical data were compared using the χ² test, and continuous data were analyzed using the 2-tailed Student t test or analysis of variant, as appropriate. Receiver operating characteristics (ROC) analysis determined performance of the scoring system in predicting overall response, and principal component analysis identified predictors of response within the scoring system. Proportions of normal and abnormal Lyon scores were compared against DeMeester scores to determine concordance and discordance, as well as performance characteristics in predicting response to therapy. In all cases, P < 0.05 was required for statistical significance. All statistical analyses were performed using IBM SPSS Statistics V.26.0 (Armonk, NY).

RESULTS

Data from 281 patients (median age 53 years, 57.7% female) were analyzed as part of the developmental cohort, 71.9% with heartburn, 13.9% with regurgitation, and 14.3% with chest pain. The median Lyon score was 6.5 (intraquartile range [IQR] 2.5–10.5, range 0.0–12.0). Medical management (255 with once daily PPI, 7 with twice daily PPI, one with H2RA) was pursued in 263 patients, while 18 patients underwent ARS. Using the Lyon score, patients could be phenotyped according to the demonstration of evidence for conclusive GERD (Figure 1). Lyon score ranges defining these phenotypes with and without conclusive GERD, and the median Lyon scores within these phenotypes are described in Table 2. These phenotypes demonstrated increasing likelihood of response to antireflux therapy as evidence of conclusive GERD increased. Symptom response was similar with both medical management (48.7%) and ARS (50.0%). Median Lyon score was higher in responders compared with nonresponders with both medical management (10.0 vs. 2.8, P < 0.001) and ARS (8.0 vs. 2.0, P = 0.031). Response was very low with functional heartburn (7.1%), increasing to 37.5% with borderline GERD and ≥ 76.6% with conclusive GERD (P < 0.001 across groups, Figure 2). The median DeMeester score was 13.3 (2.3–17.1), and scores within each phenotype are described in Table 2. On comparison of the DeMeester score (threshold 14.7) with the Lyon score, there was concordance between the 2 scores in 246 patients (87.6%); 29 patients (10.3%) had a normal DeMeester score when the Lyon score was abnormal, while only 6 patients (2.1%) had an abnormal DeMeester score when the Lyon score was normal (P < 0.001). 



On ROC analysis, the Lyon score demonstrated an AUC of 0.819 in predicting 50% symptom improvement (P < 0.001), with an optimal Lyon score threshold of 6.25 in predicting response (sensitivity 81.2%, specificity 73.4%) (Figure 3). The Lyon score performed better than any of the individual components of the score, and only AET (AUC 0.799, P < 0.001), MNBI (AUC 0.765, P < 0.001), and REs (AUC 0.764, P < 0.001) approached the AUC of the combined score (Table 4). Compared with the DeMeester score, the Lyon score was significantly superior in predicting therapeutic response in the developmental cohort (DeMeester AUC: 0.779, P = 0.019 compared with Lyon Score AUC). On principal component analysis, AET, MNBI, and RE all demonstrated high values within the component matrix (≥0.80 for each), suggesting that each of these primary metrics carries independent weight in driving response to antireflux therapy. Using the DeMeester score threshold of 14.7, sensitivity and specificity for predicting 50% response were 69.6% and 78.3%, respectively. 


Validation cohorts
Two separate validation cohorts were analyzed: a European cohort from Italy and an Asian cohort from Taiwan (Table 3). The European cohort consisted of 215 patients with heartburn studied for persisting heartburn symptoms despite PPI (median age 58.0 years, IQR 47.0–65.0 years, 60.0% female), while the Asian cohort consisted of 258 previously untreated patients (median age 47.0 years, IQR 37.0–58.0 years, 60.2% female). Consequently, proportions of patients with dominant symptoms of heartburn, regurgitation, and esophageal chest pain were different between the primary cohort and the 2 validation cohorts (Table 3). The European cohort had higher proportions with physiologic RE numbers, while the Asian cohort had numerically higher proportions of patients with AET < 4%. Proportions within subsets of numbers of REs, MNBI, RSA, and hiatus hernias were similar to the primary cohort (Table 3). In the European cohort, 198 patients were treated with single dose PPI and 17 with double dose PPI, while all patients in the Asian cohort were treated with single dose PPI; none received H2RA, and none underwent ARS. 


The median Lyon score was 2.5 (1.0–9.0) in the European cohort and 4.3 (2.5–7.6) in the Asian cohort. In addition to increasing median Lyon scores, there was a gradient of increasing proportions with 50% symptom improvement with antireflux therapy across the diagnostic categories in both validation cohorts, as well as the developmental cohort (Figure 2). The AUC for therapeutic response prediction was 0.908 (P < 0.001) in the European cohort and 0.637 in the Asian cohort (P < 0.001).

The Lyon score performed better than the major or primary components of the score in the validation cohorts (Table 4). The median DeMeester score was 9.1 (3.8–25.1) in the European cohort and 4.1 (1.4–9.2) in the Asian cohort. Sensitivity and specificity in predicting 50% treatment response were 82.5% and 94.1% in the European cohort and 23.0% and 89.0% in the Asian cohort. On ROC analysis, Lyon score gains over the DeMeester score were more evident in the Asian cohort with regurgitation predominance rather than the heartburn predominant European cohort (Table 4). Using optimal Lyon score thresholds, discordance with the DeMeester score in the European cohort was noted in 9 patients (normal Lyon score but abnormal DeMeester score in 6, and abnormal Lyon score but normal DeMeester score in 3, P < 0.001) and in the Asian cohort in 53 patients (8 and 45 respectively, P < 0.001).

When evaluating the Lyon score for 95% sensitivity (threshold of 2.75), of the 138 patients with 50% response, 131 (94.9%) had Lyon score ≥ 2.75. At 95% specificity (threshold of 11.5), 25 (18.1%) had a Lyon score ≥ 11.5 (Table 4). Within the European and Asian validation cohorts, the Lyon score threshold of 2.75 identified 83 (85.6%) and 95 (77.9%) with 50% symptom response, respectively. Only 14 (14.4%) and 15 patients (12.3%) with therapeutic response were found to have Lyon score ≥ 11.5. Furthermore, of the 118 nonresponders in the European cohort, 116 (98.3%) had Lyon score < 11.5, while in the Asian cohort, 131 of 136 nonresponders (96.3%) had Lyon score < 11.5 (Table 4).

Finally, overall, the Lyon score threshold identified 56 of 65 patients (86.2%) with therapeutic response when AET was inconclusive (4%–6%), while the DeMeester score threshold identified 44 of 65 (67.7%, P = 0.02).

DISCUSSION

The Lyon score was designed in concert with and to complement the recently introduced Lyon Consensus 2.0 criteria for conclusive GERD (4). In this retrospective study evaluating the value of the novel Lyon score, we demonstrate that the Lyon score effectively segregates symptomatic patients with typical GERD symptoms into defined and clinically meaningful phenotypes. More importantly, the Lyon score takes into account increasing confidence in the presence of conclusive GERD when multiple tests demonstrate GERD evidence, and this translates into a higher proportion with symptom improvement from antireflux therapy with increasing scores. While the DeMeester score functions well when pH-based parameters are used to predict response to antireflux therapy, the Lyon score provides added value especially in segregating nonreflux phenotypes from conclusive GERD and in predicting response with regurgitationpredominant GERD phenotypes where information from pHimpedance plays a key role in defining the presence of GERD. Furthermore, we demonstrate that the Lyon score functions well both in patients who have been initially empirically treated with a PPI and in treatment-naive patients who undergo investigation before treatment. We conclude that the Lyon score has potential to become a useful clinical tool with potential for automated calculation and prediction of symptom response, especially if validated in prospective studies.

The Lyon score is based on the concepts underlying the recently published update of the Lyon consensus (4) and assigns scores to parameters to establish or rule out the diagnosis of GERD that are available primarily from endoscopy and ambulatory reflux monitoring. The intent in designing the Lyon score was to assign weight to increasing confidence in the presence of conclusive GERD when multiple metrics demonstrate GERD evidence. Thus, each test result was given a weighted score based on a logistic regression model that ascribes confidence in predicting symptom response so that a numeric score could replace uncertainty in GERD evidence. In keeping with this paradigm, we demonstrate that phenotypes with minimal or no evidence of conclusive GERD (functional heartburn, within the disorder of gut-brain interaction spectrum) demonstrated very low proportions with symptom improvement with antireflux therapy, while phenotypes with conclusive GERD based on Lyon 2.0 principles had > 75% with symptom improvement. Intermediate phenotypes of reflux hypersensitivity and inconclusive GERD had less robust response compared with conclusive GERD, in keeping with the uncertainty of the presence of GERD or potential overlap with functional heartburn in these phenotypes but also demonstrating that GERD likely overlaps with these phenotypes.

The concept of using scoring systems to define GERD is not new. Of high clinical impact, the DeMeester score was designed to apply weight to multiple metrics from the ambulatory pH study and generate a composite score, and is widely used especially in the surgical world (19). However, reflux monitoring has advanced significantly beyond single channel 24-hour pH monitoring, and both prolonged wireless pH monitoring and pH-impedance monitoring provide advantages over pH monitoring. Wireless pH monitoring provides an estimate of day-to-day variation in acid exposure (20,21), as well as a more reliable estimate of RSA from increased symptom reporting (22). pH-impedance monitoring increases the accuracy of estimation of numbers of REs (23,24) and provides novel metrics including MNBI (16) as a measure of longitudinal damage of the esophageal epithelium due to reflux, with particular value in regurgitation-predominant reflux and complex symptomatic presentations. Furthermore, the role of a hiatus hernia in promoting reflux, as well as characterization and grading of hernias using HRM are better understood and implemented (14). In contrast to the DeMeester score which is based on pH metrics alone, the Lyon score takes all of the metrics from pH-impedance monitoring into consideration, as well as endoscopic esophagitis grade and the presence or absence of hiatus hernias. Other scoring systems have attempted to quantify symptoms, particularly the GERD questionnaire (25), reflux disease questionnaire (26), reflux symptom index (27) and GERD health related quality of life (28), all of which have modest performance characteristics in predicting conclusive GERD on objective testing.

Taking these factors into consideration, and the fact that the Lyon consensus is now established as a means for identification of conclusive GERD, the Lyon score is a concept that can advance the value of scoring systems in modern GERD diagnosis and management. Indeed, there was concordance with the DeMeester score in the majority, especially in the heartburn-predominant European patients. However, there were important areas of discordance, particularly when the DeMeester score had a normal value in the context of conclusive GERD by modern standards. Overall, the improved performance characteristics of the Lyon score over the DeMeester score, especially in regurgitationpredominant GERD in the Asian cohort, likely reflect the value of adjunctive metrics from endoscopy and pH-impedance monitoring in determining conclusive GERD. Our secondary analysis suggests that the Lyon score may have additional value in predicting response when AET is inconclusive (4%–6%), by giving weight to adjunctive metrics in improving confidence in the presence of GERD. We believe that the Lyon score may complement the recently introduced Milan score that uses HRM to estimate GERD potential on the basis of EGJ tone, EGJ morphology, EGJ performance under stress as estimated by pressure response to the straight leg raise maneuver, and esophageal body peristaltic performance (29). Indeed, the Milan score has been demonstrated to predict conclusive GERD (AET > 6.0%) with high accuracy (29) but is yet to be tested in terms of therapeutic outcome.

The differences in the results between the developmental cohort and the 2 validation cohorts provide interesting insights into ethnic and geographic differences regarding symptom evaluation, GERD phenotypes, and symptom outcome. It is well known that there are ethnic differences in GERD symptomatic presentations, indications for GERD testing, GERD evidence, and therapeutic outcome between Eastern and Western populations (30). In the Asian care pathway, there were higher proportions of patients with regurgitation and higher PPI response rates in the non-GERD cohorts compared with Western cohorts which likely contributed to the lower performance characteristics of the Lyon score. Since PPI response was not an exclusion, non-GERD foregut disorders with PPI response and regurgitation-based behavioral syndromes could have participated in the lower Asian performance. In the European cohort, PPI responders would not have been included, and the cohort only included patients with predominant heartburn. However, PPI response rates were consistently high in the conclusive patients with GERD in all cohorts. In sum, Lyon score performance characteristics were higher in the heartburn-predominant Western populations, especially those previously failing a PPI trial, which is precisely the dominant profile of patients needing invasive esophageal testing. In all instances, the performance characteristics of the Lyon score surpassed that of the individual components of the score, indicating that the coexistence of conclusive features on testing adds to confidence in a GERD diagnosis, and predicts treatment outcome.

The value of the Lyon score is demonstrated in the segregation of GERD phenotypes and prediction of outcomes in the developmental cohort as well as in 2 separate validation cohorts from different patient care delivery systems. The strengths of the Lyon score rely on ease of use and potential for reproducibility across all clinical settings, especially if an electronic or web-based system can be designed for data entry. The Lyon score provides a continuous variable that can be used to gauge severity of GERD, and potentially help the practitioner decide between a conservative medical management approach for lower scores, or a more invasive surgical or endoscopic approach when scores are high. All of the Lyon score items are available from diagnostic tests that are already part of the standard GERD evaluation (endoscopy and pH impedance monitoring), and each item has approximately 3 subscores, which makes the score quick, practical, and generalizable. This study cannot address how a DeMeester-like score based on wireless pH monitoring compares with the Lyon score based predominantly on pH-impedance metrics. The discussion of whether the value of obtaining day-to-day AET variation with wireless pH monitoring supplants value from adjunctive pHimpedance monitoring metrics is an ongoing conundrum being actively studied, both in terms of patient outcome and cost analysis. However, wireless pH monitoring is not universally available or cost-effective around the world, and pH-impedance monitoring continues to be used extensively, especially outside the United States Therefore, the Lyon score is likely to have clinical utility in these contexts.

The primary limitation of our study is that the Lyon score was tested in convenience cohorts of consecutive patients with existing data and recorded proportions with 50% symptom response rather than a prospective cohort, even though these cohorts were prospectively collected in the individual motility centers. There can also be discrepancy between symptom response and esophageal test findings, which could not be assessed because of the retrospective design. Thus, our findings need prospective and external validation. Furthermore, the presenting symptoms of patients were different between the studied cohorts, but this could also be seen as an advantage, in that the score performed well in all cohorts. This study only evaluated the Lyon score in patients with typical symptoms studied off antisecretory therapy; atypical symptoms were not included and need to be separately studied. Since the Lyon consensus now includes criteria for esophageal testing performed on PPI, it remains to be determined if the Lyon score will also predict outcome in on PPI cohorts. While functional heartburn had the lowest scores, we did not evaluate if the Lyon score can segregate different functional syndromes in an enriched cohort. We also did not test the Lyon score in a consecutive cohort of patients with erosive esophagitis where reflux monitoring is not recommended by the Lyon consensus. Wewere not able to extract information regarding specific PPIs used, doses of medications, and proportions undergoing antireflux surgery in this initial exploratory study. Future studies are planned to apply the Lyon score to specific medical and surgical management cohorts, as well as atypical GERD symptoms. Finally, the fact that the Lyon score performed well indicates that our initial attempt at weighting using logistic regression modeling was appropriate. Use of machine learning could further fine tune weighting and improve the Lyon score. Nevertheless, we feel that the current approach is an initial step toward simplifying interpretation of esophageal testing for the average clinician who manages patients with GERD. It is anticipated that with the use of sophisticated supervised machine learning algorithms and models, the concept of the Lyon score could evolve into an artificial intelligence platform that can confirm conclusive GERD and predict outcome from management.

In conclusion, we demonstrate that a simple weighted score can segregate patients with typical GERD symptoms into clinically meaningful phenotypes and can predict outcome from antireflux therapy. The Lyon score has characteristics that make it more comprehensive than the DeMeester score and has potential to impact the evaluation and management of GERD if validated in prospective studies. Future prospective studies are planned to determine if care pathways can be directed by the Lyon score, perhaps into cohorts that definitely do not need GERD management vs. those that do need antisecretory therapy and/or antireflux surgery.


CONFLICTS OF INTEREST

Guarantor of the article: C. Prakash Gyawali, MD, FRCP, FACG.

Specific author contributions: C.P.G.: study concept, data analysis, manuscript preparation and critical review. L.M.: data assimilation and analysis. B.D.R., W.W.C.: data analysis, critical review. M.W., P.W., D.C., A.R., E.S., N.d.B., C.C.: data collection, manuscript preparation. J.P.: study concept, critical review.

Financial support: This project was supported by the NIH/NIDDK R01 DK092217.

Potential competing interests: C.P.G.: Medtronic, Diversatek, Braintree, Carnot Laboratories. B.R.: Medpace, Phathom, Sanofi. W.W.C.: Sanofi, Regeneron. D.A.C.: Medtronic, Braintree, Medpace. E.V.S.: has served as speaker for AbbVie, Agave, AGPharma, Alfasigma, CaDiGroup, Celltrion, Dr Falk, EG Stada Group, Fenix Pharma, Galapagos, Johnson & Johnson, JB Pharmaceuticals, Innovamedica/Adacyte, Lilly, Malesci, Mayoly Biohealth, Omega Pharma, Pfizer, Reckitt Benckiser, Sandoz, SILA, Sofar, Takeda, Tillotts, Unifarco; has served as consultant for AbbVie, Agave, Alfasigma, Biogen, Bristol-Myers Squibb, Celltrion, Dr. Falk, Eli Lilly, Fenix Pharma, Johnson&Johnson, JB Pharmaceuticals, Merck & Co, Nestl`e, Pfizer, Reckitt Benckiser, Regeneron, Sanofi, SILA, Sofar, Takeda, Unifarco; he received research support from Bonollo, Difass, Pfizer, Reckitt Benckiser, SILA, Sofar, Unifarco, Zeta Farmaceutici. N.d.B.: speaker for: Reckitt-Benkiser, Malesci, Sofar, Dr Falk,. Advisory Board: Astra-Zeneca, Sanofi, Dr. Fali, Alfasigma. J.P.: Medtronic, Diversatek, Phathom, Takeda, Astra Zeneca, Endogastric solutions, Ethicon. L.M., M.W.W., P.V., A.R., C.L.C.: no disclosures. 

Study Highlights

WHAT IS KNOWN

~ The Lyon Consensus, recently updated, is widely used for conclusive diagnosis of gastroesophageal reflux disease (GERD).

~ Metrics from pH-impedance monitoring complement acid exposure time, endoscopic findings, and the presence/ absence of hiatus hernia in determining conclusive GERD.

~ The DeMeester score uses metrics from pH-only monitoring to identify GERD phenotypes that benefit from antireflux therapy.

WHAT IS NEW HERE

~ The novel Lyon score uses metrics and thresholds recommended by the Lyon Consensus 2.0 to characterize esophageal syndromes into meaningful phenotypes.

~ The Lyon score provides the practicing physician with a simple linear numeric score that incorporates findings from multiple esophageal tests to ascertain confidence on GERD diagnosis and to predict treatment response.

~ The Lyon score outperforms the DeMeester score in predicting typical symptom response to antireflux therapy, particularly in regurgitation-predominant GERD phenotypes.

REFERENCES

1. Yadlapati R, Gyawali CP, Pandolfino JE, et al. AGA clinical practice update on the personalized approach to the evaluation and management of GERD: Expert review. Clin Gastroenterol Hepatol 2022;20(5): 984–94.e1.
2. Katz PO, Dunbar KB, Schnoll-Sussman FH, et al. ACG clinical guideline for the diagnosis and management of gastroesophageal reflux disease.Am J Gastroenterol 2022;117(1):27–56.
3. Roman S, Gyawali CP, Savarino E, et al. Ambulatory reflux monitoring for diagnosis of gastro-esophageal reflux disease: Update of the Porto consensus and recommendations from an international consensus group. Neurogastroenterol Motil 2017;29(10):1–15.
4. Gyawali CP, Yadlapati R, Fass R, et al. Updates to the modern diagnosis of GERD: Lyon consensus 2.0. Gut 2024;73(2):361–71.
5. Demeester TR, Johnson LF, Joseph GJ, et al. Patterns of gastroesophageal reflux in health and disease. Ann Surg 1976;184(4):459–70.
6. Rogers BD, Patel A, Wang D, et al. Higher esophageal symptom burden in obese subjects results from increased esophageal acid exposure and not from dysmotility. Clin Gastroenterol Hepatol 2020;18(8):1719–26.
7. Lundell LR, Dent J, Bennett JR, et al. Endoscopic assessment of oesophagitis: Clinical and functional correlates and further validation of the Los Angeles classification. Gut 1999;45(2):172–80.
8. Yadlapati R, Masihi M, Gyawali CP, et al. Ambulatory reflux monitoring guides proton pump inhibitor discontinuation in patients with gastroesophageal reflux symptoms: A clinical trial. Gastroenterology 2021;160(1):174–82.e1.
9. Patel A, Sayuk GS, Gyawali CP. Parameters on esophageal pH-impedance monitoring that predict outcomes of patients with gastroesophageal reflux disease. Clin Gastroenterol Hepatol 2015;13(5):884–91.
10. Rogers BD, Valdovinos LR, Crowell MD, et al. Number of reflux episodes on pH-impedance monitoring associates with improved symptom outcome and treatment satisfaction in gastro-oesophageal reflux disease (GERD) patients with regurgitation. Gut 2021;70(3):450–5.
11. Patel A, Wang D, Sainani N, et al. Distal mean nocturnal baseline impedance on pH-impedance monitoring predicts reflux burden and symptomatic outcome in gastro-oesophageal reflux disease. Aliment Pharmacol Ther 2016;44:890–8.
12. Patel A, Sayuk GS, Gyawali CP. Prevalence, characteristics, and treatment outcomes of reflux hypersensitivity detected on pH-impedance monitoring. Neurogastroenterol Motil 2016;28(9):1382–90.
13. Patel A, Sayuk GS, Kushnir VM, et al. GERD phenotypes from pHimpedance monitoring predict symptomatic outcomes on prospective evaluation. Neurogastroenterol Motil 2016;28(4):513–21.
14. Rengarajan A, Gyawali CP. High-resolution manometry can characterize esophagogastric junction morphology and predict esophageal reflux burden. J Clin Gastroenterol 2020;54(1):22–7.
15. Visaggi P, Del Corso G, Gyawali CP, et al. Ambulatory pH-impedance findings confirm that grade B esophagitis provides objective diagnosis of gastroesophageal reflux disease. Am J Gastroenterol 2023;118(5):794–801.
16. Martinucci I, de Bortoli N, Savarino E, et al. Esophageal baseline impedance levels in patients with pathophysiological characteristics of functional heartburn. Neurogastroenterol Motil 2014;26(4):546–55.
17. Gyawali CP, Rogers B, Frazzoni M, et al. Inter-reviewer variability in interpretation of pH-impedance studies: The Wingate consensus. Clin Gastroenterol Hepatol 2021;19(9):1976–8.e1.
18. Gyawali CP, Patel A. Esophageal motor function: Technical aspects of manometry. Gastrointest Endosc Clin N Am 2014;24(4):527–43.
19. Johnson LF, DeMeester TR. Development of the 24-hour intraesophageal pH monitoring composite scoring system. J Clin Gastroenterol 1986; 8(Suppl 1):52–8.
20. Pandolfino JE, Richter JE, Ours T, et al. Ambulatory esophageal pHmonitoring using awireless system.AmJ Gastroenterol 2003;98(4):740–9.
21. Penagini R, Sweis R, Mauro A, et al. Inconsistency in the diagnosis of functional heartburn: Usefulness of prolonged wireless pH monitoring in patients with proton pump inhibitor refractory gastroesophageal reflux disease. J Neurogastroenterol Motil 2015;21(2):265–72.
22. Prakash C, Clouse RE. Value of extended recording time with wireless pH monitoring in evaluating gastroesophageal reflux disease. Clin Gastroenterol Hepatol 2005;3(4):329–34.
23. Bredenoord AJ. Impedance-pH monitoring: New standard for measuring gastro-oesophageal reflux. Neurogastroenterol Motil 2008;20(5):434–9.
24. Bredenoord AJ, Weusten BL, Timmer R, et al. Addition of esophageal impedance monitoring to pH monitoring increases the yield of symptom association analysis in patients off PPI therapy. AmJ Gastroenterol 2006; 101(3):453–9.
25. Jones R, Junghard O, Dent J, et al. Development of the GerdQ, a tool for the diagnosis and management of gastro-oesophageal reflux disease in primary care. Aliment Pharmacol Ther 2009;30(10):1030–8.
26. Dent J, Vakil N, Jones R, et al. Accuracy of the diagnosis of GORD by questionnaire, physicians and a trial of proton pump inhibitor treatment: The diamond study. Gut 2010;59(6):714–21.
27. Belafsky PC, Postma GN, Koufman JA. Validity and reliability of the reflux symptom index (RSI). J Voice 2002;16(2):274–7.
28. Velanovich V, Vallance SR, Gusz JR, et al. Quality of life scale for gastroesophageal reflux disease. J Am Coll Surg 1996;183(3):217–24.
29. Siboni S, Sozzi M, Visaggi P, et al. Manometric screening of patients with laryngopharyngeal reflux using the novel Milan Score. Gastroenterology 2024;56:S175.
30. Sifrim D, Roman S, Savarino E, et al. Normal values and regional differences in oesophageal impedance-pH metrics: A consensus analysis of impedance-pH studies from around the world. Gut 2020;70(8):1441–9.

Назад в раздел