Predictive Value of Endoscopic Findings for Pathologic Acid Exposure in Patients without Reflux Esophagitis or Hiatal Hernia

 Predictive Value of Endoscopic Findings for Pathologic Acid Exposure in Patients without Reflux Esophagitis or Hiatal Hernia

Vinod Kumar Singhal *1, Faris Dawood Alaswad 2, Dr Adil Mohamed Sulaiman3, Omer Elfaroug Amin Mohammed4, Dr Nufra Senopher5, Sanjay Kumar Sureen6


1) Vinod Kumar Singhal- Consultant General Surgeon, Prime Hospital, Dubai

2) Faris Dawood Alaswad-Consultant General Surgeon, Department of Surgery, Gladstone Hospital, Perth, Australia.

3) Dr Adil Mohamed Sulaiman- Specialist Surgeon, Department of General Surgery, Prime Hospital, Dubai, UAE.

4) Omer Elfaroug Amin Mohammed-Resident Specialist Surgeon, Department of General Surgery, Prime Hospital, Dubai, UAE.

5) Dr Nufra Senopher -Specialist ENT Surgeon, Department of General Surgery, Prime Hospital, Dubai, UAE.

6) Sanjay Kumar Sureen- Specialist Orthopedic Surgeon, Department of Orthopedic Surgery, Prime Hospital, Dubai, UAE.


*Correspondence to: Vinod Kumar Singhal consultant General Surgeon, Prime Hospital, Dubai.


Copyright.

© 2025 Vinod Kumar Singhal This is an open access article distributed under the Creative Commons Attribution   License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original  work is properly cited.

Received: 29 July 2025

Published: 01 Aug 2025

DOI: https://doi.org/10.5281/zenodo.16794093

Abstract

Background/Aims:

Diagnosing gastroesophageal reflux disease (GERD) remains difficult in patients who present with reflux-like symptoms but lack evidence of reflux esophagitis or hiatal hernia (HH) on endoscopy and do not respond to proton pump inhibitor (PPI) therapy. This study aimed to determine whether specific endoscopic findings, particularly the gastroesophageal flap valve (GEFV) grade, could predict pathologic acid exposure (PAE) in this patient group and to develop an endoscopic prediction model for PAE.

Methods:

A total of 578 patients with reflux symptoms but no reflux esophagitis or HH were initially screened. After exclusions, 494 patients who underwent both upper endoscopy and 24-hour pH monitoring were analyzed. The presence of GEFV grades, esophageal metaplasia (ESEM), and chronic atrophic gastritis (CAG) were assessed, and their associations with PAE were evaluated through multivariable logistic regression to create a predictive model.

Results:

Among the 494 patients, the most frequent symptom was chest discomfort (42.3%), followed by globus sensation (31.8%), dysphagia (7.9%), and heartburn (7.7%). PAE was observed in 43 patients (8.7%). Multivariate analysis showed that a higher GEFV grade was significantly associated with PAE (p<0.001), while the presence of CAG was inversely related to PAE risk (p=0.005). An endoscopic prediction model based on GEFV and CAG demonstrated moderate discriminative ability (AUROC 0.705), with a cutoff score of 12.0 providing 44.0% sensitivity and 84.0% specificity.

Conclusions:

In patients without reflux esophagitis or HH, a loosened GEFV increases the likelihood of PAE, whereas CAG appears protective. However, due to the low sensitivity of the endoscopic model, additional reflux monitoring is warranted when GERD is suspected despite normal endoscopy.

 Predictive Value of Endoscopic Findings for Pathologic Acid Exposure in Patients without Reflux Esophagitis or Hiatal Hernia

Introduction

Diagnosing GERD in patients who lack visible endoscopic markers such as reflux esophagitis or HH presents a clinical challenge, especially when these patients also fail to respond to PPI therapy. Although ambulatory reflux monitoring is the gold standard for confirming GERD, its limited availability and patient discomfort hinder widespread use. In contrast, upper endoscopy is widely accessible and helps identify anatomic features of the esophagogastric junction (EGJ) such as the GEFV, which may contribute to reflux pathogenesis.

GERD is multifactorial, but HH is often central to its mechanism. However, patients without HH may still experience reflux due to transient lower esophageal sphincter relaxations (TLESRs) or weakened anatomical barriers. The EGJ comprises several components, including the lower esophageal sphincter (LES), crural diaphragm, and the anatomical flap valve. These components together function as the anti-reflux barrier. Although the absence of HH often indicates a less disrupted EGJ, abnormalities such as a loosened GEFV can still impair its function.

This study evaluates the predictive significance of endoscopic findings, specifically the GEFV grade and the extent of CAG, for detecting PAE in patients without overt reflux esophagitis or HH. A mathematical model for predicting acid reflux based on endoscopic parameters was developed and validated.

 

Methods

Patient Selection

Between June 2011 and February 2015, 578 adult patients with symptoms suggestive of GERD underwent upper endoscopy and 24-hour pH monitoring within a one-month interval. After excluding patients with previous upper GI surgeries or pneumatic dilation (n=44), and those with visible reflux esophagitis, 494 patients were included in the final analysis. Ethical approval was obtained, and the need for informed consent was waived due to retrospective data review.


Endoscopic and Clinical Evaluation

Demographic and clinical data, including chief symptoms and endoscopic images, were reviewed. Endoscopic assessments focused on:

  • GEFV Grade: Assessed based on flap valve opening width relative to endoscope diameter.
  • CAG: Graded using the Kimura classification.
  • ESEM: Identified visually using Prague criteria (segment length >5 mm).

All patients discontinued acid-suppressive therapy at least seven days prior to pH monitoring. A pH probe was positioned 5 cm above the LES to monitor acid exposure. PAE was defined as pH <4 for more than 4.2% of the 24-hour monitoring period.


Statistical Analysis

Univariate and multivariate logistic regression models were used to identify associations between endoscopic findings and PAE. A prediction model was developed from a 70:30 split of the data into derivation and validation sets, with ROC curve analysis used to evaluate model performance. SAS and R software were used for analysis.

 

Figure1: Representative endoscopic images illustrating the grading of the gastroesophageal flap valve (GEFV):
(A) Grade 0 – Tight valve with an opening width ≤1 cm.

(B) Grade 1 – Mildly loosened valve with an opening width between 1.0 and <1.5 cm.

(C) Grade 2 – Moderately loosened valve with an opening width between 1.5 and <2.0 cm.

(D) Grade 3 – Severely loosened valve with an opening width >2.0 cm.

 

Results

Patient Characteristics

The mean age of the 494 patients was 53 years, and 64.8% were female. The most commonly reported symptom was chest discomfort (42.3%), followed by globus (31.8%), dysphagia (7.9%), and heartburn (7.7%). PAE was detected in 8.7% of the subjects.

 

Endoscopic Findings

  • GEFV Grades: Grade 0 (≤1 cm opening) was most common (65.0%), followed by Grade 1 (26.5%), Grade 2 (6.5%), and Grade 3 (2.2%).
  • CAG: Present in 60.8% of subjects, with 17.6% classified as severe.
  • ESEM: Detected in 17.4% of patients.


Associations with Pathologic Acid Exposure

Multivariate regression revealed:

  • A higher GEFV grade significantly increased the odds of PAE. Grade 3 had an OR of 14.66 (p<0.001).
  • CAG was inversely associated with PAE. Severe CAG had an OR of 0.22 (p=0.005).
  • ESEM had a marginal association (p=0.061).


Prediction Model for PAE

A predictive model was created using GEFV and CAG grades:

  • Score Formula: A mathematical formula integrating GEFV and CAG grades was established.
  • Best Cutoff Score: 12.0 yielded an accuracy of 81%, sensitivity of 44%, and specificity of 84%.
  • AUROC: The final model had an AUROC of 0.705 (95% CI: 0.619–0.790), indicating moderate discriminatory ability.

A nomogram was developed to visually estimate PAE probability using total points derived from GEFV and CAG grades.

 

Discussion

This study demonstrates that the structural condition of the GEFV and the presence of atrophic gastritis are independent predictors of PAE in patients lacking visible endoscopic signs of GERD or HH. A loosened GEFV weakens the mechanical barrier, thereby increasing susceptibility to reflux, even in the absence of HH. Conversely, chronic atrophic gastritis may reduce acid secretion, offering a protective effect against reflux.

Despite these findings, the low sensitivity of the endoscopic prediction model limits its standalone clinical use. The model’s utility may lie in identifying patients at higher risk for PAE, who would benefit from further diagnostic testing such as 24-hour impedance-pH monitoring, especially when GERD remains a strong clinical suspicion.

Notably, this study employed objective measurements for flap valve grading rather than subjective classifications, improving reproducibility. However, the retrospective design, reliance on image review, and ethnic homogeneity (all Korean subjects) limit generalizability. Additionally, ESEM assessment was limited to ultrashort segments, which may not represent the full spectrum of Barrett’s esophagus.

 

Conclusions

In patients with reflux symptoms but no endoscopic evidence of reflux esophagitis or HH, a loosened GEFV significantly increases the risk of pathologic acid exposure, while atrophic gastritis appears to reduce this risk. Although the derived endoscopic prediction model offers moderate accuracy, it has limited sensitivity, underscoring the need for further reflux testing when GERD is clinically suspected.

 

References

1.  Hershcovici T, Fass R. 2013; Step-by-step management of refractory gastresophageal reflux disease. Dis Esophagus. 26:27–36. DOI: 10.1111/j.1442-2050.2011.01322.x. PMID: 22309405.

2. Boeckxstaens G, El-Serag HB, Smout AJ, Kahrilas PJ. 2014; Symptomatic reflux disease: the present, the past and the future. Gut. 63:1185–1193. DOI: 10.1136/gutjnl-2013-306393. PMID: 24607936. PMCID: PMC4078752.

3. Kahrilas PJ, Boeckxstaens G, Smout AJ. 2013; Management of the patient with incomplete response to PPI therapy. Best Pract Res Clin Gastroenterol. 27:401–414. DOI: 10.1016/j.bpg.2013.06.005. PMID: 23998978. PMCID: PMC3761380.

4. Herregods TV, Bredenoord AJ, Smout AJ. 2015; Pathophysiology of gastroesophageal reflux disease: new understanding in a new era. Neurogastroenterol Motil. 27:1202–1213. DOI: 10.1111/nmo.12611. PMID: 26053301.

5. van Herwaarden MA, Samsom M, Smout AJ. 2000; Excess gastroesophageal reflux in patients with hiatus hernia is caused by mechanisms other than transient LES relaxations. Gastroenterology. 119:1439–1446. DOI: 10.1053/gast.2000.20191. PMID: 11113064.

6. Dodds WJ, Dent J, Hogan WJ, et al. 1982; Mechanisms of gastroesophageal reflux in patients with reflux esophagitis. N Engl J Med. 307:1547–1552. DOI: 10.1056/NEJM198212163072503. PMID: 7144836.

7. Bredenoord AJ, Pandolfino JE, Smout AJ. 2013; Gastro-oesophageal reflux disease. Lancet. 381:1933–1942. DOI: 10.1016/S0140-6736(12)62171-0. PMID: 23477993.

8. Curcic J, Roy S, Schwizer A, et al. 2014; Abnormal structure and function of the esophagogastric junction and proximal stomach in gastroesophageal reflux disease. Am J Gastroenterol. 109:658–667. DOI: 10.1038/ajg.2014.25. PMID: 24589669.

9. Park CH, Kim KO, Baek IH, et al. 2014; Differences in the risk factors of reflux esophagitis according to age in Korea. Dis Esophagus. 27:116–121. DOI: 10.1111/j.1442-2050.2012.01417.x. PMID: 23009198. PMCID: PMC6543060.

10. Kim GH, Song GA, Kim TO, et al. 2008; Endoscopic grading of gastroesophageal flap valve and atrophic gastritis is helpful to predict gastroesophageal reflux. J Gastroenterol Hepatol. 23:208–214. DOI: 10.1111/j.1440-1746.2007.05038.x. PMID: 18289353..