INTRODUCTION
Gastroesophageal reflux disease (GERD) is one of the most common digestive disorders worldwide. With an estimated prevalence between 8% and 33%, GERD significantly impacts quality of life and represents an important economic burden on healthcare systems.1 The modern definition of GERD requires objective evidence of pathological reflux, consisting of conclusive findings on endoscopy or multichannel intraluminal impedance and pH monitoring (MII-pH), in the presence of troublesome symptoms. When the diagnosis is uncertain because of inconclusive findings, the Lyon Consensus 2.0 proposes evaluating for adjunctive metrics supportive of GERD.
One of these is the presence of a hiatal hernia (HH), a proximal protrusion of the gastric cardia and fundus through a lax diaphragmatic hiatus.2 Common in Western countries, HH is recognized as a key factor compromising the integrity of the esophagogastric junction (EGJ) barrier and facilitating gastroesophageal reflux.3,4
Despite the known association between HH and GERD, the impact of HH size remains a subject of debate. While older studies have reported that large HHs are associated with increased acid exposure time (AET), esophagitis, and decreased lower esophageal sphincter (LES) competence,5,6 newer evidence suggests weak or inconsistent correlations between HH size and reflux parameters assessed by MII-pH, implying that HH presence may be more important than size.7 This is potentially related to significant variations in the definition of HH size, which is sometimes classified as type II or III EGJ morphology on high-resolution manometry (HRM), and at other times by numeric size based on HRM, barium radiography, or endoscopy. The Lyon Consensus proposes a threshold of 2 cm to objectively define the presence of HH on the basis of physiological evidence showing that this degree of separation significantly compromises anti-reflux function.8
Few studies have evaluated reflux parameters using this stricter HH definition, and even fewer have simultaneously incorporated MII-pH measurements and endoscopic findings, which allow for better characterization of the reflux burden. Clarifying this relationship could optimize diagnostic stratification and guide individualized management strategies. Therefore, the present study aimed to evaluate the correlation between HH size measured on HRM and objective GERD parameters on endoscopy and MII-pH in symptomatic patients comprehensively evaluated at a referral center.
MATERIALS AND METHODS
Design
This was a retrospective observational study of a consecutive cohort of symptomatic patients with suspected gastroesophageal reflux who presented to the gastrointestinal physiology department between January 2018 and June 2025. Adult patients over 18 years of age with typical or atypical GERD symptoms who had undergone HRM, MII-pH, and upper endoscopy within ≤6 months of each other were included. HH was identified, and its size was measured using HRM.
Patients with incomplete studies due to technical failure or data omission, a history of upper esophagogastric surgery, or the use of medications that altered esophageal motility (metoclopramide, anticholinergics, and opioids) were excluded. Owing to the retrospective nature of the study, the study was performed with a waiver of informed consent. The study protocol was approved by the Ethics Committee (FM-CIE-0471-25).
Data Collection
Eligible patients were identified through an institutional procedure registry and subsequently verified through a detailed review of medical records. HRM was performed after a minimum fast of 6 hours, using a solid-state catheter with 36 circumferential sensors (Medtronic), following the Chicago Consensus v4.0 protocol.9 Measurement of HH size was made during the resting “landmark” phase of the HRM study, which was characterized by visible separation between the LES and the crural diaphragm (CD) impression. The distance between the lower border of the LES at expiration and the inspiratory CD impression was considered as the size of the HH (Fig.
1).
MII-pH was performed using a Versaflex Z catheter placed 5 cm above the proximal margin of the manometrically identified LES, after discontinuation of proton pump inhibitors, H2 antagonists, antacids, and prokinetics for at least 7 days, in accordance with international recommendations.2 The MII-pH tracings were processed using Reflux Reader Software v6.1.1.11 for analysis. The Los Angeles (LA) classification was systematically applied during endoscopy for characterization of erosive esophagitis.10 HRM and MII-pH analysis were performed independently by two evaluators, of whom the senior evaluator had >5 years of experience.
Operational Definitions
Using the thresholds proposed by the Lyon Consensus, HH was defined as a separation of ≥2 cm between the LES and the CD measured by HRM and classified into three categories: <3 cm (small), 3–5 cm (medium), and >5 cm (large). The absolute HH size was also measured. The primary endpoints were the proportions of patients with AET >6% on MII-pH and conclusive GERD findings on endoscopy (LA grades B, C, and D esophagitis, peptic stricture, or Barrett’s esophagus) in relation to the HH size.
Secondary endpoints included the DeMeester score (pathologic when ≥14.7), total reflux episodes (TRE), acid-reflux and non-acid reflux episodes, and mean nocturnal baseline impedance (MNBI, pathologic when <1500 ohm). Clinical demographic variables (age, sex, and comorbidities), typical symptoms (heartburn, regurgitation, and chest pain), and atypical symptoms (dysphagia, cough, and belching) reported by the patients were also recorded.
Statistical Analysis
The sample size was calculated considering a clinically relevant effect size of 0.3, statistical power of 80%, and significance level of 5%, which allowed us to establish a minimum required size of 440 patients.
Continuous variables were assessed for normality using the Shapiro–Wilk test and were described as mean ± standard deviation or median with interquartile range (IQR), as appropriate. Categorical variables were expressed as absolute frequencies and percentages. The Kruskal–Wallis test was used to compare continuous variables.
Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Correlational analysis was performed comparing AET, MNBI, and TRE to absolute HH size. Sensitivity analysis was performed to compare MNBI between groups after excluding patients with esophagitis and Barrett’s esophagus, which are known to be associated with low baseline impedance.
To assess the independent association between HH size and reflux burden, multiple linear regression models were used for each outcome: AET, DeMeester score, and TRE. The models were adjusted for sex, age, symptom phenotype (perceptive [heartburn, chest pain], transit [regurgitation], and others [atypical]), and the presence of a hypomotility disorder (ineffective esophageal motility and absent contractility) because of their potential role as confounding factors. Regression coefficients are presented with 95% confidence intervals (CIs) and P-values.
Statistical significance was set at P < 0.05. All analyses were performed using STATA version 16.1 (StataCorp).
RESULTS
Of the 3298 HRM studies reviewed within the study period, 2779 were excluded because MII-pH or complementary upper-endoscopy data were unavailable within the pre-specified 6-month time window, and 63 studies were excluded for a history of esophagogastric surgery. A cohort of 337 patients with HH (median age, 62 years; IQR, 54–69 years; 71.2% female) was eventually analyzed. Of these, 114 (33.8%) had small HHs, 117 (34.7%) had medium HHs, and 106 (31.4%) had large HHs.
The dominant symptom was heartburn (63.8%), and 20.0% of patients also had dysphagia. A comparison cohort of 119 patients (median age, 58 years; IQR, 48-68 years; 72.3% female) without HH was also analyzed (Table 1).
The pathological AET criterion (>6%) was met in 19.3% of the non-HH cohort and 57.8% of the HH cohort, with a distribution that increased proportionally with HH size: 41.2% in small hernias, 51.3% in medium hernias, and 83.0% in large hernias (P < 0.001). Progressive and statistically significant increments in the median AET were found between groups, with median values of 2.3% (IQR, 0.45–4.95) for no HH and 4.7% (IQR, 0.9–8.9), 6.0% (IQR, 3.1–13.7), and 13.2% (IQR, 7.3–22.5) for small, medium, and large hernias, respectively (P < 0.001, Kruskal–Wallis test). Similarly, the prevalence of erosive esophagitis increased with HH size, from 7.6% in cases with no HH to 14.9%, 17.9%, and 43.4% in cases with small, medium, and large HH, respectively (P < 0.001) (Fig.
2). In contrast, no significant differences were observed in the frequency of peptic strictures (P = 0.263) or the presence of Barrett’s esophagus (P = 0.108) between the cohorts (Table 2).
Relationship to Additional Parameters on pH Impedance Monitoring and Endoscopy
The DeMeester score showed a progressive and consistent increase between groups, with median values of 8.7 (IQR, 2.8–20.3) in patients with no HH and 16.4 (IQR, 5.7–32.6), 22.9 (IQR, 12.1–50.3), and 49.2 (IQR, 29.6–94.5) in patients with small, medium, and large HH, respectively (P < 0.001). The proportion of patients with a pathological DeMeester score was 39.5% in those with no HH, increasing incrementally as HH size increased (54.4%, 69.2%, and 88.7%; P < 0.001). The total number of reflux episodes was 41 (IQR, 24.5–65.5) in patients with no HH and 93 (56–148), 131 (86–184), and 160 (115–250) in patients with small, medium, and large HH, respectively (P < 0.001) (Table 2).
Pathological TRE (>80 episodes) increased proportionally with the presence and size of the HH (P < 0.001) (Fig. 2). The total number of acid-reflux episodes showed a similar increase, with median values of 14 (IQR, 6–31.5), 53 (IQR, 29–104), 92 (IQR, 39–152), and 139 (IQR, 75–217), respectively, in each group, and statistically significant differences in both the supine and upright fractions (P < 0.001).
In contrast, the number of non-acid-reflux episodes did not significantly differ between the groups (P = 0.193) (Table 2). Finally, MNBI was significantly lower in patients with larger hernias and showed statistically significant differences between the groups when pathological (<1500 ohm) (P < 0.001) (Table 2 and Fig. 3).
Sensitivity Analysis
We performed two sensitivity analyses. First, MNBI was compared between groups after excluding patients with erosive esophagitis and Barrett’s esophagus, and the findings showing progressively lower MNBI with larger hernias remained consistent (P < 0.001) (Fig. 2).
Next, AET was compared between patients with HH without dysphagia as a reported symptom. The AET differences among the three HH subsets remained statistically significant, whereas AET remained within the physiological range regardless of dysphagia in patients without HH (Table 2).
Correlational Analysis
When absolute HH size was correlated with AET, TRE, and MNBI values, a significant positive correlation was observed between HH size and AET (Spearman rho = 0.49, P < 0.001) and TRE (Spearman rho = 0.53, P < 0.001), and a negative correlation was observed between HH size and MNBI (Spearman rho = -0.46, P < 0.001) (Fig. 4).
Multivariable Regression Analysis
In a multivariable regression model adjusted for age, sex, symptom phenotype, and presence of esophageal hypomotility, HH size remained independently associated with reflux burden (Table 3). Relative to no HH, small, medium, and large HHs were associated with progressively higher rates of AET (β, 3.86 [95% CI, 1.12 to 6.60], 6.71 [95% CI, 3.94 to 9.49], and 14.50 [95% CI, 11.60 to 17.41], respectively), higher DeMeester scores (β, 14.26 [95% CI, 4.35 to 24.18], 24.47 [95% CI, 14.37 to 34.58], and 55.66 [95% CI, 45.09 to 66.22], respectively), and more TRE (β, 82.76 [95% CI, 52.72 to 112.81], 114.02 [95% CI, 83.71 to 144.33], and 147.71 [95% CI, 116.13 to 179.29], respectively); all associations were statistically significant. Esophageal hypomotility was associated with a higher AET, whereas older age was associated with a lower TRE.
Subgroup Analysis of LES-CD Separation of ≤2 cm vs. >2 cm
Patients with LES-CD separation >2 cm had a significantly greater reflux burden and lower mucosal integrity than those with LES-CD separation ≤2 cm (Supplementary Table 1). Specifically, median AET in patients with LES-CD separation >2 cm was 7.45 (IQR, 2.75–14.75) vs. 2.3 (IQR, 0.45–4.75) in patients with LES-CD separation ≤2 cm, while median DeMeester scores were 28.1 (IQR, 11.2–54.25) vs.
9.0 (IQR, 2.8–19.85) and median TRE were 130 (IQR, 75–191) vs. 42 (IQR, 25–67), respectively (all P < 0.001). In contrast, median MNBI was significantly lower in patients with LES-CD separation >2 cm than in those with LES-CD separation ≤2 cm (1040 ohm [IQR, 520–2290 ohm] vs.
2325 ohm [IQR, 1420–3080 ohm]; P < 0.001).
DISCUSSION
In this retrospective observational study that evaluated the probability of conclusive GERD in relation to HH size,2 we demonstrated a consistent gradient between HH size and objective GERD parameters on MII-pH monitoring and erosive esophagitis. As the size of the HH increased, there was corresponding statistically significant increase in AET, frequency of esophagitis, DeMeester score, and TRE, especially acid-reflux episodes, in both the supine and upright positions, accompanied by a marked decrease in MNBI. Significant positive correlations were identified between the absolute HH size and AET as well as TRE, while absolute HH size showed a negative correlation with MNBI.
In contrast, the presence of peptic strictures and Barrett’s esophagus did not show statistically significant differences, although they were numerically higher in larger hernias. We conclude that, in addition to the presence of HH, the size of the HH matters in determining confidence in diagnosing conclusive GERD.
The findings of this study suggest that HH size is a useful clinical marker for estimating the probability of conclusive GERD when integrated with objective physiological metrics from MII-pH testing, including AET, DeMeester score, and MNBI, particularly when these metrics are inconclusive. Importantly, this association remained significant after multivariable adjustment, supporting HH size as an independent determinant of reflux burden. From a pathophysiological perspective, our findings are consistent with available data that suggests HH progressively weakens the anti-reflux barrier as a consequence of greater separation between the LES and the CD.11 Pandolfino et al.8 demonstrated that separations ≥2 cm significantly impair CD function and increase susceptibility to prolonged episodes of acid reflux, and our findings support this by showing a higher reflux burden as the HH size increased.
The progressive reduction in MNBI even after adjusting for presence of esophagitis and Barrett’s esophagus is consistent with existing reports demonstrating that the presence of HH is an independent factor for compromised mucosal integrity, especially in the context of biliary reflux.12 In clinical practice, our data suggest that an HH >5 cm indicates a high probability of significant anti-reflux barrier impairment and abnormal AET; thus, in patients with a large symptomatic HH, ambulatory reflux monitoring may not necessarily change diagnosis or management decisions. These findings may also have practical and translational relevance. In subgroup analysis, LES-CD separation >2 cm, which was used as a surrogate marker of more advanced EGJ disruption and potentially corresponded to endoscopic Hill grades III–IV, was associated with a higher reflux burden and lower MNBI, suggesting that EGJ anatomy may help estimate GERD severity when esophageal physiologic testing is not available.
While large hernias can generate artifacts in HRM metrics from angulation of the EGJ or obstruction from concurrent paraesophageal hernia13 leading to a paradoxically lower AET, our cohort of large HHs was associated with a higher reflux burden than smaller hernias. Our findings are consistent with the results of two studies demonstrating very high acid-reflux metrics in giant sliding HH in comparison with small HHs.14,15 An Italian study also demonstrated that patients with type 3 EGJ morphology corresponding to large HHs had higher DeMeester scores than those with type 2 (small HH) and type 1 (no HH) morphology.16 Other studies have suggested that while AET in the presence of hernia is higher than that in the absence of hernia, the difference in AET between type 2 and type 3 morphology is not statistically significant.7,17 In our study, large HHs (>5 cm) were analyzed separately from HHs 3–5 cm in size, which provided evidence of a progressive increase in acid burden as the HH size increased. The higher prevalence of esophagitis in large hernias was consistent with the higher reflux burden in this subset of patients.
In contrast, the absence of significant differences in Barrett’s or peptic strictures suggests that abnormal acid exposure of any magnitude could predispose to these findings in patients at risk for these conditions. The higher frequency of Cameron ulcers exclusively in large hernias is consistent with their mechanical pathophysiology, which is associated with mucosal entrapment and repeated trauma to the hiatus.3
A limitation of our study is that the paraesophageal component of HH was not systematically recorded, since we relied on manometric HH detection, in which paraesophageal hernias cannot be reliably identified. However, HRM measurement of HH has been demonstrated to have high sensitivity and specificity in comparison with actual measurements during laparoscopic anti-reflux surgery.18 Barium studies were not consistently performed in our patient cohort. Since the presence of a paraesophageal component may be associated with obstructive symptoms, we performed a sensitivity analysis by stratifying patients according to the presence or absence of reported dysphagia.
Consistently, a lower AET was observed in the three HH subsets in patients with dysphagia. However, when restricting the analysis to patients without reported dysphagia, the difference in AET among the three HH subsets remained statistically significant. Finally, although this study included a large cohort evaluated using HRM, MII-pH, and endoscopy, its retrospective single-center design, large number of excluded patients, and absence of longitudinal follow-up limited the ability to infer causality or assess progression to chronic complications.
In addition, the retrospective design may have introduced selection bias, since most excluded patients underwent HRM primarily for the evaluation of transit symptoms (dysphagia) rather than reflux symptoms, and therefore did not undergo ambulatory reflux monitoring.
In conclusion, HH size is significantly associated with the conclusive diagnosis of GERD, with larger hernias demonstrating greater acid exposure, lower mucosal integrity, and a higher probability of erosive esophagitis. These findings reinforce the need to incorporate objective measurement of HH size into the comprehensive evaluation of patients with suspected GERD and in personalized therapeutic decision-making.